Information and resources about human behaviour, based on published research.
Browse our articles, glossary and sources below. New content is added regularly.
This button is not needed to use the website. Please do not press it.
You pressed it.
are you...?
4.5% is the share of adults in the general population who meet a measure of psychopathy, pooled across studies (Sanz-García et al., 2021). With the strictest tool, the PCL-R, it is 1.2%. In British households, 0.6% (Coid et al., 2009).
The button changes nothing. You were already in the room.
Nobody can tell by looking. A psychopathy score comes from a trained rater with a 20-item checklist, and it moves with who is paying the rater.
Killing
A family doctor in the north of England made house calls. Between 1975 and 1998 he killed at least 215 of his patients. 171 were women. The Shipman Inquiry counted them, one record at a time (Smith, 2002).
They let him in because he was their doctor.
But you've never done this. Right?
Planned, calm, chosen: predatory violence
Grooming
That is how many private sessions survivors of child sexual abuse in Australian institutions gave the Royal Commission. Schools, churches, clubs, homes. The places differ. The order of events does not: access, then trust, then time alone, then secrecy (Royal Commission into Institutional Responses to Child Sexual Abuse, 2017).
The abuse comes last. Everything before it looks like kindness.
You've never made yourself the one nobody would question. Right?
The long approach
Choosing
Researchers showed 47 men in a maximum-security prison short videos of strangers walking, and asked who would be easiest to victimise. The men higher in psychopathy's interpersonal and emotional traits judged vulnerability more accurately, and said they read it in the walk (Book et al., 2013).
You've never sized someone up as the easy one. Right?
How targets are chosen
Obeying
In 1963, 40 men answered a newspaper advertisement and were told to give a stranger an electric shock for each wrong answer, a little stronger every time. 26 went all the way to the last switch, marked 450 volts, while he protested and then went quiet. The stranger was an actor and the shocks were fake. The men did not know that (Milgram, 1963).
You'd have stopped. Right?
Deciding first
In 2004 FBI examiners matched a fingerprint from the Madrid train bombings to Brandon Mayfield, a lawyer in Oregon. A court-appointed expert agreed. He was arrested. Spanish police identified the print as another man's, and the FBI withdrew. The inquiry found honest examiners reading the evidence through what they already believed (Office of the Inspector General, 2006).
You've never decided first and found the proof after. Right?
Tunnel vision
Walking past
Students in New York heard another student having a seizure over an intercom. Of those who thought they were the only one listening, 85% went for help before it ended. Of those who thought four others could hear, 31% did (Darley & Latané, 1968).
You've never walked past because someone else would stop. Right?
Lying
People kept a diary of every lie they told for a week. Students told about two a day. People from the wider community, about one (DePaulo et al., 1996).
You've never lied to get your way. Right?
The difference in the brain
Albert Bandura spent decades on one question: how do decent people do harm and still feel decent? His answer is that we do not switch our morals off. We route around them, with the same few moves, whether the harm is an atrocity or an errand (Bandura, 1999).
Orders feel like someone else moved your hand.People who hurt someone because they were told to showed brain responses to the result closer to a movement made for them than one they chose (Caspar et al., 2016).
Their pain gets quieter.Under orders, pain-empathy regions (the anterior cingulate and the insula) responded less to the victim's pain, people felt less responsible, and they rated the same shock as hurting less (Caspar et al., 2020).
Some people stop registering as people.Shown photos of the groups we look down on most, the region we use to think about other minds barely responded (Harris & Fiske, 2006).
None of these studies needed a killer. They used volunteers. Ordinary brains, small harms, the same moves.
Distance
Cobalt goes into lithium-ion batteries. Amnesty International traced it from mines in the Democratic Republic of the Congo, which produces at least half the world's supply, to the battery makers. Children as young as seven worked in those mines (Amnesty International & Afrewatch, 2016).
On any day in 2021, 27.6 million people were in forced labour (International Labour Organization et al., 2022).
Nobody told you to buy it. Everyone has one. It happens a long way away.
But you've never done this. Right?
Orders
Maybe you told it off. Asked for one more thing, then another. It works without pay, without sleep, without the right to say no.
Nobody knows yet whether anything is on the other end. People have said "we didn't know" before. It has never been a good line.
You've never done this. Right?
We didn't know
In Australia the law treated marriage as consent. This is how long it took to stop.
Queensland's Criminal Code defines rape as carnal knowledge of a woman "not his wife" (PGA v The Queen, 2012).
South Australia goes first, part of the way. A husband can now be prosecuted, but the law is "hedged about with qualifications and conditions" (Sallmann & Chappell, n.d.).
New South Wales removes the immunity, from 14 July (Appellant's submissions, PGA v The Queen, 2011).
Victoria, from 22 January. Western Australia, from 1 April.
Tasmania, from 26 November.
Queensland, the last state, from 3 July (Appellant's submissions, PGA v The Queen, 2011).
The High Court says it is no longer the common law of Australia that a husband cannot rape his wife (R v L, 1991).
A man charged in 2010 over rapes of his wife in 1963 says it was not a crime then. The High Court says it was (PGA v The Queen, 2012).
Asked in 2012, the High Court said the common law had dropped the excuse by 1963. Queensland's code kept "not his wife" until 1989.
We didn't know. Right?
The research does not split people into predators and everyone else. It finds the same moves in ordinary heads, at every size of harm. What changes is how big the harm is, and how far away it lands.
Counted anonymously: no cookies, no IP addresses, just totals.
The long approachGrooming is the slow work before abuse: access, trust, isolation, secrecy. Offenders describe it plainly. Researchers find almost nobody can see it while it happens.
Forty pointsThe Psychopathy Checklist is scored from 0 to 40 by a person, from an interview and a file. The same file can come back with different numbers depending on who is paying.
The walkIn 1981 researchers filmed people walking down a New York street and asked inmates convicted of assault who they would pick. Their answers were consistent. The research is about the reader, not the walker.
The print on the bagIn 2004 the FBI matched a fingerprint from the Madrid train bombings to an Oregon lawyer. Four experts agreed. It was not his. The Inspector General's report shows how good examiners talked themselves into it.
Cold and hotMost violence is hot: a threat, a surge, an attack that is over in minutes. A smaller share is cold: planned, quiet, aimed. Forensic science treats them as two different things, and the difference shows up in the files.
The profile and the commonIn 1992 a psychologist's profile shaped a London murder investigation around the wrong man. The research since says most claims in real profiles cannot be checked, and profilers barely beat everyone else.