In 1962, the American television show Candid Camera aired a segment called "Face the Rear." Actors stepped into a lift and turned to face the back wall, and unsuspecting passengers often turned too. It's funny, memorable and widely shared. It is also an everyday demonstration, not a study: edited for laughs, with no count of people who didn't turn. So what do the controlled studies say?
Three lines and a room full of confident strangers
laboratory findingStarting in 1951, Solomon Asch asked participants to say which of three lines matched a reference line. The answer was obvious, but the other people in the room were confederates who sometimes gave the same wrong answer. Across critical trials, participants went along with the group roughly a third of the time. About three-quarters conformed at least once, while around a quarter never did.
The detail that rarely makes the headline: when just one confederate gave the correct answer, conformity dropped sharply. A 1996 meta-analysis by Rod Bond and Peter Smith found conformity in Asch-type studies declined in the United States over the decades and tended to be higher in more collectivist cultures.
Study limit: mostly young male students in 1950s America, judging a trivial task
Obedience, and what the archive later showed
laboratory finding · contested detailsStanley Milgram's 1963 study at Yale reported that 26 of 40 men continued to the maximum 450-volt level on a fake shock generator when an experimenter told them to. It became shorthand for "ordinary people will follow orders." Less famous are Milgram's many variations: obedience fell when the experimenter gave instructions by phone or when other "teachers" refused.
In Behind the Shock Machine (2012), Gina Perry reviewed the archive and described departures from the script and participants who doubted the shocks were real. Jerry Burger's 2009 partial replication, which stopped at 150 volts for ethical reasons, still found high rates of continuing. The pressure is real; the precise percentages are less certain than textbooks imply.
The bystander story that grew in the retelling
"38 witnesses watched Kitty Genovese die and did nothing."
The 1964 New York Times account made the case a parable of urban apathy.
Exaggerated numbers, real lab effect
A 2007 article by Rachel Manning, Mark Levine and Alan Collins argued the number of witnesses and their passivity were overstated. Separately, John Darley and Bibb Latané's 1968 lab studies did show that people helped less, and more slowly, when they thought others were present.
Then came real-world footage. A 2020 study led by Richard Philpot analysed CCTV recordings of public conflicts in Amsterdam, Lancaster and Cape Town. In about nine out of ten incidents, at least one bystander intervened, and the chance someone helped rose with the number of people present. Each person may feel less responsible, but a crowd contains more potential helpers.
The prison study as a cautionary tale
weak evidenceThe 1971 Stanford Prison Experiment, led by Philip Zimbardo, was stopped after six days and is often cited as proof that roles turn people cruel. In 2019, Thibault Le Texier published an archival analysis in American Psychologist arguing that guards were coached toward harshness. The BBC Prison Study by Alex Haslam and Stephen Reicher (2002) produced very different group dynamics. The safest reading: it was a dramatic demonstration, not a controlled experiment.
Social influence is often about whose voice counts. One dissenting voice can change a room.
Observation prompt
In a queue or waiting room, notice what happens after one person takes out a phone, stands up or checks the time. Count how often others follow and how often they don't. Both counts matter.
What to take away
Groups clearly influence us, but the influence is conditional. Allies, the authority's presence, culture, and how people interpret the situation all change the outcome. The classic studies remain valuable as starting points, especially when read alongside the critiques and real-world data that came after them.