The Wason Test

Last week Grubby again won the prize of recognition for having genius puzzle-solving skills—this time without giving any of the other readers a chance to buzz-in first. Rachel claimed to have gotten the right answer too, and Bobby et al. winked-winked a hint that the puzzle wasn't as hard as the one from two weeks ago. Though we're not dealing with scientifically suitable sample sizes here, what we have is confirmation of the Wason Test.

Many of you may have noticed that those two puzzles were, in the abstract, the same. Here's the mapping:

1st puzzle	2nd puzzle
if one side of a card shows a vowel	if a person is drinking alcohol
then the card's other side must show an even number	then the person must be 21 or older
card showing E	person with face-down ID drinking beer
card showing K	person with face-down ID drinking water
card showing 4	21-or-older person drinking unknown drink
card showing 7	underage person drinking unknown drink

But also as many of you may have noticed, the first puzzle feels harder than the second. But it doesn't just feel harder: it is harder, being as how most people get it wrong. I got it wrong. My excuse is I rushed my answer and failed to think it through, but that's the point: the first puzzle is deceptively hard. You've got to slow down and think it through or else you're likely to make a mistake. Whereas, with the second puzzle it's easy to figure out you need only check the people who may be underage and drinking alcohol. There's nothing deceptive about it.

The curiousness of the Wason Test is why one puzzle is hard while the other is easy. If both puzzles are identical mathematically then it seems they ought to have identical difficulty. But these puzzles are night-and-day different in difficulty. What causes the discrepancy?

Perhaps that question is the hardest puzzle of all.