Why does the same logic puzzle get easy once you wrap a story around it?

Only the content. In the abstract phrasing, with letters and numbers, correct answers run around ten to twenty percent. Put exactly the same logical problem as a social rule, that anyone drinking alcohol has to be over a certain age, and most people solve it without effort. Nobody hesitates over whether to check the drinker. The logical structure is identical and only the wrapper moved.

The task itself is easy to describe and remarkably hard to solve. You are shown four cards and a rule of the form "if one side is such, the other side is such". Then you are asked which cards must be turned over to check whether the rule is broken.

In the abstract phrasing, with letters and numbers, correct answers run around ten to twenty percent. Almost everyone turns the card that would confirm the rule, and almost nobody turns the one card that might refute it. That is confirmation bias in its cleanest form.

And then comes the interesting part. Put exactly the same logical problem as a social rule, that anyone drinking alcohol has to be over a certain age, and most people get it right immediately. Nobody wonders whether to check the drinker. The logical structure is identical, and only the content and context changed.

The careful conclusion is that our reasoning does not run general rules of logic but leans on knowledge tied to a context. The bolder reading holds that we have a dedicated cheater detection mechanism, and it is disputed: other researchers explain the gap simply, as familiarity with the situation. Either way the practical lesson is the same. A problem you are stuck on is worth restating as a story.