● A free public tool  ·  Built by Poptani Labs  ·  No sign-up  ·  Your data never leaves your browser
The Human Operating System · 5-minute read

The bystander effect: why nobody helps when everyone is watching

A crowd of people gathered in the street to watch a Punch and Judy show
Image: 'A crowd of people have gathered in the street to watch a Pun Wellcome V0040135.jpg' by Wikimedia Commons, CC BY 4.0, via Wikimedia Commons

On the night of 13 March 1964, a 28-year-old woman named Kitty Genovese was attacked outside her apartment building in Kew Gardens, Queens, New York. Two weeks later, the New York Times reported that 38 respectable citizens had watched a killer stalk and stab her for more than half an hour — and that none of them had called the police. The story became one of the most famous in psychology. It also turned out to be mostly wrong. But the research it inspired is real, and the numbers from the laboratory are startling. This is a research-based retelling, drawn from the original studies and the later investigations that corrected the record.

The experiment: a seizure on the intercom

Two New York psychologists, John Darley and Bibb Latané, read the Times story and asked a different question from everyone else's: what if the witnesses were not callous, but caught in a situation that made helping less likely? In 1968 they tested the idea at New York University. Students sat alone in booths, talking over an intercom about university life, and were told the discussion involved a certain number of other students. One of the voices — a recording — then began to stammer, call for help, and appear to have a seizure: "I-er-um-I think I-uh-I need-er-if-if somebody would-er-give me a little h-help-uh-er-er-er-er-er c-could somebody-er-er-help-er-er-seizure-er-". The measure was simple: did the student leave the booth to find the experimenter before the fit ended? When students believed they were the only listener, 85% went for help. When they believed two others could hear, 62% did. When they believed four others were listening, only 31% did. Darley and Latané called the mechanism diffusion of responsibility: when others might act, each person feels less obliged to, and each can quietly assume someone else has already gone for help. A second mechanism, pluralistic ignorance, makes it worse in ambiguous situations — people look at each other for cues, see nobody reacting, and conclude nothing is wrong.

The famous story was mostly wrong

Here is the twist that textbooks took decades to catch up with. In 2007, psychologists Rachel Manning, Mark Levine and Alan Collins went back to the archive and published their findings in American Psychologist: they found "no evidence for the presence of 38 witnesses, or that witnesses observed the murder, or that witnesses remained inactive". Only about half a dozen people testified at the trial as having seen any part of the attack, and nobody had seen the whole thing — it happened in the dark, in separate stages, largely out of view. A neighbour named Robert Mozer heard Kitty's screams, opened his window and shouted "Leave that girl alone!" — and the attacker fled. Another neighbour, Sophia Farrar, ran out, found Kitty wounded in an alley, and held her, whispering that help was on the way, until the ambulance came. People did call the police. In 2016 the Times itself admitted its original article had "grossly exaggerated the number of witnesses and what they had perceived". The number 38 may have come from a police synopsis listing almost 50 interviewees grouped into 38 entries — a count that was passed along, possibly over lunch between a police commissioner and a Times editor, and printed as 38 people who watched a murder. The story that launched a field of psychology was, in its famous form, a counting error.

The effect is real — but narrower than the legend

None of that makes the bystander effect itself false. A 2011 meta-analysis by Peter Fischer and colleagues, covering around 100 studies, confirmed the effect on average — but found it much weaker, and sometimes reversed, when the situation is clearly dangerous, when a perpetrator is present, or when bystanders can see each other's concern. And the cameras have added a corrective of their own: in a 2019 study of 219 filmed public conflicts in three countries, at least one bystander intervened in 90.9% of cases, and each extra bystander raised the odds that someone did. The honest summary is narrower than the legend: a crowd does not make people heartless. It makes ambiguous situations harder to read, and it spreads the feeling of responsibility so thin that each person can believe someone else is handling it. When the emergency is obvious, people usually help — and they help more, not less, when others are around to help with them.

The Human Angle

The bystander effect is not a verdict on human nature; it is a warning about ambiguity. If you are the person in trouble, don't shout at the crowd — point at one person and give them a job: "You, in the blue shirt, call an ambulance." If you are the bystander, assume nobody else has called, and act as if you are the only one who noticed. The 38 witnesses were mostly a myth. The diffusion of responsibility is not — but it breaks the moment one person decides the job is theirs.

Related on Poptani Labs: Nobody noticed: the spotlight effect, and why embarrassment lies to you, Daniel Kahneman and the two systems that run your mind, Decision fatigue: why judges said no before lunch. More in this category: The Human Operating System.

Sources: Murder of Kitty Genovese (Wikipedia), Research Methods in Psychology — Darley & Latané 1968 (Lumen Learning), Bystander Effect: Does a Crowd Stop People Helping?

← All posts