The easiest person to fool

About this video

Months have passed. Zara's bean plant has grown tall and has beans on it, and her notebook is full: observations, three hypotheses, predictions, a lamp-and-heater test in which the plant leaned to the lamp, and notes from friends who tried it too. She has found out that plants grow towards light. But she has learned something bigger: a way of finding out that works for almost any question about the world. It isn't a list of facts; it is a set of habits. Look carefully and write it down. Guess in ways you could test, and say what the guess predicts before you look. Test fairly: one change at a time, compare with a control, let chance choose the groups, hide the labels, try many times. Share your work so others can check it and do it again. And change your mind as much as the evidence calls for, and no more.

In 1974 the physicist Richard Feynman gave a talk to students finishing their studies. "The first principle," he told them, "is that you must not fool yourself - and you are the easiest person to fool." Every tool in the course is a way of not fooling yourself. The control group stops you thinking "it got better" means "it worked". Random groups stop you choosing the keenest without noticing. Blind tests stop your expectations changing the result, as with Clever Hans. Replication stops one lucky result from fooling everyone.

You don't need a laboratory or a white coat: a question, a notebook, and the habit of asking "how could I find out?" and then "how could I be wrong?" This video goes with the Thinking Like a Scientist lesson of Science Level 1 on Knaph.

Transcript

1. 0:00Months later, Zara's notebook is full. She learned more than why plants lean.
2. 0:05Look carefully. Guess in ways you can test. Say what it predicts.
3. 0:11Test fairly. Share your work. Change your mind as much as the evidence says.
4. 0:17Feynman said, you must not fool yourself. And you're the easiest person to fool.
5. 0:23Control groups, coin tosses, blind tests, doing it again. Each one stops a fooling.
6. 0:29Science is a way of not fooling yourself. Anyone can do it.