The vision test that wasn't
In 1951, psychologist Solomon Asch began a series of famous experiments at Swarthmore College. He recruited male college students for what they were told was a simple "vision test." The setup was straightforward: a group of people sat in a classroom and were shown two cards. One card had a single vertical line, the "reference line." The second card had three vertical lines of different lengths. The task was to state, out loud, which of the three lines on the second card was the same length as the reference line.
The answer was always obvious. However, only one person in the room was a genuine participant. The others, typically between seven and nine people, were confederates, or actors, who had agreed in advance what their answers would be. The real subject was placed at or near the end of the row, so they would hear most of the others' answers before giving their own. The experiment involved 18 trials for each participant. For the first few trials, everyone gave the correct answer. Then, on 12 of the trials, known as "critical trials," the confederates all gave the same, clearly incorrect answer. The experiment measured whether the real subject would stick to their own perception or conform to the group's error.
The power of the group
The results were unexpected. Across all the critical trials, participants conformed to the incorrect majority about 37% of the time. While a majority of individual answers were correct, the social pressure had a significant effect. 75% of all participants conformed at least once during the experiment. A control group, where participants performed the task alone without confederates, showed an error rate of less than 1%. This demonstrated that the task itself was not ambiguous. When interviewed afterward, most participants who conformed said they did so to fit in or because they were afraid of being ridiculed, a phenomenon called normative influence.
Asch also tested several variations. He found that the size of the majority mattered, but only up to a point. Conformity peaked with a majority of three confederates; adding more people did not significantly increase the effect. The most dramatic variation involved the presence of a "true partner." When just one of the confederates was instructed to give the correct answer alongside the real participant, conformity dropped to only 5%. This showed that having a single dissenting ally was enough to shatter the majority's social power. When participants were allowed to write their answers in private, conformity almost vanished completely.