Today I’d like to talk about an important phenomenon in psychology that pertains to how we do research. Before we start, I’d like to give you a small test. You see, I cannot give you the test after I’ve discussed it with you. Please humor me and spend a minute taking the test in below. No information about you will be collected or stored.
Did you do it already?
No?
…
I’ll wait.
…
Let’s look at this cute picture of a snoozing cat while we push the explanation of the text further down on the computer screen to avoid people with large monitors to accidentally catch a glimpse of the solution before they have a chance to try the test.
Sorry for the subterfuge.
You see, I was testing your confirmation bias. That was the point of the test. But it would have somewhat given it away if I admitted it upfront. Did you correctly guess the rule? Most people get it wrong. Usual guesses are “the next number is +2 of the last”, “only even numbers”, or something similar. The trick is that, in order to verify that you have the right rule, you actually have to make a guess that you think will fail. Something like \(3, 2, 1\). Only by using both positive and negative guesses can you correctly deduce the rule.
This test was devised by P.C. Wason and is know as the 2-4-6 task. And it brilliantly illustrates our bias to looking for confirmation. We have a tendency to look for support for our theories. It’s likely we’ll find it. But if we don’t also look for evidence that we are wrong, we’ll never get to the truth.
This is widely applicable. Consider a person you don’t like all that much. Whenever they do something at odds with your values, you notice. Because it confirms you hypothesis that they are a Bad Person. You notice when they are rude to the bus driver, when they don’t let passengers off the bus before they enter it, when they talk on speakerphone on the bus1. You don’t notice when they do something that goes against your hypothesis. Like giving up their seat for the elderly (30+), shouting at the bus driver when they don’t open the door for an introverted person who wants to get off, or putting their backpack on their lap instead of in the seat next to them like it’s some sort of valuable object that carries more worth than other people.
It’s easy to forget once you get into the thick of research, but our goal is to look for evidence that goes against our hypothesis. The idea being that if we try our hardest to falsify a hypothesis, and fail, it’s evidence that it might be true.
Footnotes
A capital offence in most parts of the world.↩︎