The Asch conformity experiments

A psychologist ran an experiment that still unsettles people who read about it. He seated a person in a room with a group of others, all of whom were secretly actors, and showed everyone a simple task: match the length of one line to one of three others. The answer was obvious. But the actors, one by one, gave the same wrong answer out loud, without hesitating. When it came to the real participant's turn, having heard everyone before them choose an answer their own eyes knew was wrong, many went along with the group. About three quarters of participants conformed to the obvious falsehood at least once, and about a third of all the answers on those trials went with the group, though about a quarter of people never went along at all. They doubted their own clear perception rather than stand alone against a unanimous group. The result held up across many later versions, though the exact rate varies by culture and era.

People will deny what they can plainly see rather than stand alone against a confident, unanimous group, and it happens below the level of conscious courage. This is the hidden failure mode of many meetings and hiring panels: once the first strong opinion is voiced, especially by someone senior, the others fall in line, not because they were persuaded but because disagreeing alone feels wrong. The safeguards are structural. Gather people's honest views privately before any group discussion, and have the most senior person speak last, because the moment they speak first, you stop getting judgements and start getting echoes. The flip side, that genuinely independent judgements aggregate remarkably well, is the wisdom of crowds.

Rebel Ideas

Matthew Syed, 2019

On why diverse, independent thinking beats confident consensus, and how groups silence it.