I solved it in the usual way. I find it hard to believe 98% *couldn't*... in the sense that I think what stands in the way for most of them is lack of exposure to this puzzle type in the first place, much less the drill of solving such. I'm sure a significant quantity of those who wouldn't know where to begin without guidance wouldn't have much difficulty if they were given a bit of guidance (setting up the classic grid, etc.)
But how many people would find a reason to bother taking the time?
I'm sure a lot more of the planet could figure it out if they were given a few pointers about writing down (or drawing) all facts, that it's equally important of what things aren't/can't be as what they are, and that bounds are limiting factors. But on the other hand a good portion of the world are as dumb as stumps. There is a critical period where if people don't use modus ponens and modus tollens that they become psychologically incapable of it.
Given the statement: If George Washington cut down the cherry tree then he has an ax.
Modus Ponens:
If A then B. A is true therefore B is true.
He cut down the tree, therefor he has an Ax.
Modus Tollens:
If A then B. Not B therefore not A.
He has no ax, therefore he didn't (couldn't have) cut down the cherry tree.
An interesting case is the statement "he has an ax" alone, does not guarantee that he cut down he tree - B does not guarantee A (I don't remember the formal way to express that) That may seem obvious, but, according to my college propositional calculus teacher at the time, it's unknown as to why that case isn't true from a formal mathematical point of view.
A coworker was telling me about the critical period a while back. They did a study with adults never exposed to any form of formal or informal reasoning, inductive logic, or deductive logic that went something like this: All farmers have a tractor. Joe is a farmer. Does Joe have a tractor? Answer: I don't know, I don't know him. Repeat: All farmers
( ... )
Yeah, I noticed that too. My first response was to immediately count how many animals were mentioned in the list first, just in case the answer was "nobody".
But if one considers "Who owns the fish?" to be as much a part of the instructions as any of the rest of it, yes, we do know there's a fish.
I had a similar problem with a Mensa puzzle book I got a while back. The lack of detail and limitations in the scenarios made it so I could come up with alternative but equally plausible (or not: inserting concepts like Star Trek transporters) answers, particularly since many of them were based on inductive reasoning. So yeah, none of them have a fish: the Norwegian's cat ate it. Prove me wrong, I dare you! Nehh!!! :P
Comments 12
Reply
Reply
Reply
But how many people would find a reason to bother taking the time?
--Ember--
Reply
Reply
--Ember--
Reply
Modus Ponens:
If A then B. A is true therefore B is true.
He cut down the tree, therefor he has an Ax.
Modus Tollens:
If A then B. Not B therefore not A.
He has no ax, therefore he didn't (couldn't have) cut down the cherry tree.
An interesting case is the statement "he has an ax" alone, does not guarantee that he cut down he tree - B does not guarantee A (I don't remember the formal way to express that) That may seem obvious, but, according to my college propositional calculus teacher at the time, it's unknown as to why that case isn't true from a formal mathematical point of view.
A coworker was telling me about the critical period a while back. They did a study with adults never exposed to any form of formal or informal reasoning, inductive logic, or deductive logic that went something like this: All farmers have a tractor. Joe is a farmer. Does Joe have a tractor? Answer: I don't know, I don't know him. Repeat: All farmers ( ... )
Reply
Reply
But if one considers "Who owns the fish?" to be as much a part of the instructions as any of the rest of it, yes, we do know there's a fish.
--Ember--
Reply
Reply
Leave a comment