Peter Cathcart Wason at University College London. Dr. Wason pioneered the “Psychology of Reasoning,” an inquiry into the mistakes people make on a consistent basis. He coined the term “confirmation bias” to describe the tendencies he documented after running a series of experiments. One of Dr. Wason’s most famous experiments can beat it – or at least safeguard against it – by developing a simple habit that is counterintuitive for most people. The experiment, carried out in 1960, was known as the “2-4-6” task. It worked like this: • The subjects were shown a set of three numbers: 2-4-6 • The subjects were told a certain rule determines the sequence • The subjects were asked to try to figure out what the rule was For the subjects of the experiment, the 2-4- 6 task looked like a logic test. They were encouraged to propose different sets of numbers in order to test a hypothesis as to what the “rule” was that generated the 2-4-6 result. The majority of subjects in the test followed a pattern that looked like this: • They intuited the pattern was a rise of even numbers • They would offer guesses like “8-10-12” and “14-16-18” • The experimenter would say “correct” multiple times • The subject would confidently guess the rule not only underscores the presence of confirmation bias, it shows how you
The subjects of the experiment tended to be absolutely confident in their final guesses – that the rule was a pattern of rising by two or starting with an even number and adding two each time. But that was wrong: The actual rule was, “three rising numbers in sequence.” The initial 2-4-6 was a deceptive pattern. It could have been any three numbers. A guess of 3-47-104 would have worked just as well.
The human brain wants to reinforce the views it already has. If emotion is involved, the tendency gets even stronger.
The 2-4-6 task was remarkable in what it showed via the common pattern of behavior in the test subjects. • Most of the subjects started with a hunch that the 2-4-6 pattern was “rising by two each time” • Most of the subjects then offered guesses in line with their “rising by two” theory • And most of the subjects then offered their opinion with a high degree of confidence – and were likely shocked and surprised when told they were wrong! The funny thing here is, the subjects could have challenged their own hunches at any point. There was no penalty for offering a series of numbers to which the experimenter said “no, that’s not it.” All a subject would have needed to test his hunch would have been to try a series of numbers that didn’t conform to the rising-by-two rule, just to see what happened.
American Consequences 49
Made with FlippingBook Learn more on our blog