So we saw in Cog Sci today the difference between propositional and procedural statements. I got a friend to help me out with this one. For example, let's consider the following problem:
(the answer and more under the cut)
My friend looked at it, then looked at me and said angrily that he knows he's seen this before. I don't doubt he has. I know I've seen the nine-dot problem at least once every two years in the past fifteen years of my life, and every time I see it again, I know I've seen a solution but I completely forget what that solution was.
(I often wish I were like this with books so that I could enjoy them all over again but for a book it takes a much, much longer time before I can reread it and feel the same discovery-like sensation of having never read it before. Not being a psychologist, I'd still bet it has something to do with episodic versus semantic memory. If you really get 'sucked' into a book, it's like you're living the character so it's more like episodic memory rather than the semantic fact-like memory of "oh, here's the solution to a nine-dot problem; how grand. Carry on then".
tl;dr barring traumatic brain injury, it's probably more difficult to get rid of an episodic memory.)
I can give a clue to solve this problem, and that clue is "think outside the box".
..
...
.......
I don't know about you but given that clue, I was no more likely to solve the problem, and neither was my friend. So I think it's pretty useless as far as clues go. It's interesting that many people report the same feeling and it's been proven as well through testing that the clue "think outside the box" doesn't make you statistically more likely to solve the problem.
I can give another clue to solve this problem, and that clue is:
Without spoiling the answer, it's been shown that now, a solution is not so difficult. Studies have shown that you are statistically more likely to solve it with this clue rather than the clue "think outside the box". And yet in a way, this is exactly the same sort of thing - drawing a bigger box outside and around the 'box' that you think is formed by the nine dots kind of forces you by the picture to think outside of that 'box'. The statement "think outside the box" and the act of drawing the box are in a way equivalent to each other. But only one has been shown to improve success on this puzzle, so they're obviously not equivalent.
This is mostly to illustrate the difference between procedural and propositional facts. A propositional fact is some sort of knowledge-based guide. It's like knowing how Superman flies. But the important thing to note is that even if you knew how Superman flies, it doesn't give you the power to fly.
Only some of this information is easy to program into a computer by means of some sort of algorithm. All you have to do is tell the computer what you're doing; the computer has a perfect reproductive and accurate memory and therefore just sort of swallows things up and regurgitates when needed. It can even perform tasks based on rules like code. But giving the computer, say, the ability to solve a PDE does not mean you've enabled the computer to figure out when to use a PDE unless there's something in the code that explicitly calls, I dunno, use_pde(now). Similarly, giving a computer the equations of motion for a computer-like body to shake hands does not mean the computer will spontaneously get up and shake your hand. Or that it can even be trained to do that. Programming into a computer the ability to be trained, it seems, is the best we can come to giving a computer procedural knowledge, but that's still propositional knowledge at its heart.
Procedural knowledge is less knowing factual knowledge and more the ability to learn a skill. We do this so often it's almost second nature - gleaning a new skill from either factual knowledge or from some other skill, or maybe from analogy, or imitation, or any number of things that might enable us to learn.
What does this have to do with CSC236?
Very little, I've just had it on the brain for awhile. In fact I'm only really bringing this up to try and explore what propositional knowledge and propositional logic is, and I'm trying to see if exploring what it isn't makes it any easier. I don't actually think it does.
Maybe at the very least, putting all of this down in a slog entry after having put my friend through the trouble and seen it in lecture will be a good way of getting the nine-dot problem down in my semantic and episodic memory so that I'll never struggle with this stupid problem again? But it still hasn't taught me anything about insight problem solving that isn't propositional so the bigger problem is still outstanding.
(By the way the solution I was thinking of is
Now you can see how the clue "thinking outside the box" can be interpreted, but now rereading the propositional clue almost makes it feel like a bad pun.)
No comments:
Post a Comment