This neat exercise featured on the New York Times takes a few seconds to play and includes a neat set of examples of how confirmation bias impacts government policy and corporate America.

Basically, you are presented with 3 numbers in a sequence, and asked to guess the rule that governs the sequence. You can enter in any 3 numbers you like, and the system will tell you whether or not your sequence follows the rule or not. When you are ready to guess, you enter it in but you receive no second chances. Apparently 78% of people make a guess without getting a single ‘no’–and most get the rule wrong.

The example in the NY Times is ‘2, 4, 8’. A number of possible rules could come to mind–must contain multiples of 2, or even numbers, or that the number doubles the one before it. The actually rule in this case is even simpler: the number must be larger than the one before it, meaning that ‘4, 8, 16’ works, but so does ‘1, 10, 3593’.

They don’t mention it in the article, but this exercise can adapted to teach hypothesis testing. Used in class, you can put the sequence on the board and have students suggest other sequences, which you then judge as either following or not following the rule. They have to use this information to come up with the right answer.

If this sounds familiar, it may be because one of the very first entries on this blog was about the board game Zendo, which does precisely this, only with physical pieces rather than numbers. I still use Zendo on day 1 of my methods class, and find it a really useful tool for teaching a variety of methodological skills. This numerical version is a great, easy activity to pull out for a quick fix on helping students with their logical thinking.