On the whole, we’re pretty rational beings – even if behavioural scientists love to trick us
This is i on Science, a subscriber-only newsletter from i. If you’d like to get this direct to your inbox, every single week,If you read a lot of pop-science books during the 2010s you might have come away with the impression that humans are pretty silly things.
We also get taken in by “framing effects”. For instance, you might say to people that a novel disease will kill 600 people and offer them two possible responses. In the first, a well-tested but only moderately successful drug which will definitely save 200. The second is an untested long shot with a one-third chance of saving all 600 but a two-thirds chance of saving none. People reliably say, by a margin of about 3:1, that they would choose the first option.
Most people say 99 per cent, but the answer is “you don’t know.” If you test a thousand people, and only one of them actually has the disease, then your test will correctly identify that person, but also wrongly diagnose about 10 healthy people. In that scenario, if you got a positive test, there’d only be about a 10 per cent chance you were unwell. You need to know how likely it was that you had the disease before you took the test. Bayes’ theorem tells you how to do that.
An example uses another classic demonstration of human irrationality, the Wason selection task. If you give someone four cards. They have either a shape or a number on one side, and an animal or a person on the other. You show them these four:And you say to them that if a card shows a number on one side, the other side shows an animal. You can turn over two cards to check if that’s true.