The validity of game theory?

[xisnotx]

I've heard of game theory, but I've never actually studied it in depth. The way I see it, game theory is as valid as any other theory..in short, not very. I understand that you can try to analyze and predict what people are going to do in certain situations, but that's all it will be..a prediction (perhaps more reliable than others..but still ultimately flawed). Game theory,I think, assumes that people will, more often than not..make a choice based on a limited amount of variables...or that it can somehow quantify all the variables that people will consider when making a decision..even those that are unconscious.

Can it actually "predict human behavior"? Or is it "just a guess...perhaps more accurate than other systems..but ultimately just as flawed as those systems?"

Can, or does, game theory quantify the human element?

If not, then why bother? Just for fun? For profit?

And..if enough people understood game theory..wouldn't it become invalid since if enough people are doing the same thing then they are in effect not thinking for themselves, and are therefore nothing than a (good) flawed system?

 

[Rasofy]

I don't think we can ever get 100% efficiency using game theory but I believe in many cases we can get very close, e.g., when you wanna buy something with a good discount, almost always the best strategy will be act not very interested in buying the item. Unless the product is very scarce or the seller is mentally unstable, the buyer has nothing to lose acting this way because it is on the seller's interest to sell his product. The buyer may not get any discount at all, but as a general rule he won't lose anything. Not many things to be considered there.
But things get tricky in complex cases like the prisoner's dilemma.

Two suspects are arrested by the police. The police have insufficient evidence for a conviction, and, having separated the prisoners, visit each of them to offer the same deal. If one testifies for the prosecution against the other (defects) and the other remains silent (cooperates), the defector goes free and the silent accomplice receives the full one-year sentence. If both remain silent, both prisoners are sentenced to only one month in jail for a minor charge. If each betrays the other, each receives a three-month sentence. Each prisoner must choose to betray the other or to remain silent. Each one is assured that the other would not know about the betrayal before the end of the investigation. How should the prisoners act?

http://en.wikipedia.org/wiki/Prisoner's_dilemma

 

[Lark]

I've heard of game theory, but I've never actually studied it in depth. The way I see it, game theory is as valid as any other theory..in short, not very. I understand that you can try to analyze and predict what people are going to do in certain situations, but that's all it will be..a prediction (perhaps more reliable than others..but still ultimately flawed). Game theory,I think, assumes that people will, more often than not..make a choice based on a limited amount of variables...or that it can somehow quantify all the variables that people will consider when making a decision..even those that are unconscious.

Can it actually "predict human behavior"? Or is it "just a guess...perhaps more accurate than other systems..but ultimately just as flawed as those systems?"

Can, or does, game theory quantify the human element?

If not, then why bother? Just for fun? For profit?

And..if enough people understood game theory..wouldn't it become invalid since if enough people are doing the same thing then they are in effect not thinking for themselves, and are therefore nothing than a (good) flawed system?

I'm just interested in how you came to this view about theories and theorising, why dont you think any of them are valid and if you dont think it is why is T one of your identified functions? I tend to think that a lot of the theories I read about ARE valid, there are some exceptions but they dont invalidate the theory automatically, especially not established theories such as game theory.

Do you consider a theory to be an artifice or invention which if it is consciously adopted and deployed by enough people is fact/reality rather than theory? I tend to believe theories are observed facts arranged coherently into a knowledge base, some people have knowledge and they can benefit of it and some dont have it but may unconsciously or informally acts as though they do. The difference I believe is one of what I would describe as "fluency".

Game theory as I understand it is that people will co-operate until one person is seen to breech that unwritten agreement and compete/act for their self-interest, then the others involved will act competitively also, that's my understanding but I could be mistaking it for theorising about gift relationships by Marcel Maus and reciprocal obligations.

 

[Magic Poriferan]

If you are simply trying to use game theory to predict how a real person is actually going to behave, you are using it wrong.

When one applies game theory, they determine from the start kind of behavior they are expecting from the players and then chart out what would happen as a result. In the prisoner's dilemma, it is assumed from the get-go that players are going to act in a way that will bring the largest amount of gain for themselves within their scope of awareness. We then examine what happens as a result. We all know a human being won't necessarily act like this, and there are entirely different studies for figuring out how people really do tend to behave, not to mention competitions to see who can actually create the most successful long-term strategy.

Game theory is an analytical method mainly for determining outcomes, not how people will behave in the first place.

 

[Jonny]

I would suggest you make an attempt to understand something before you pass judgment. Also, make note of the distinction between the colloquial usage of the term theory and its use in the scientific world. Theories are not simply conjectures or guesses about the nature of things, but tried and tested models which have solid foundations and have yet to be disproven.

Also, MP is correct in his assertion that those who attempt to use a theory in ways which are outside its scope are simply "using it wrong." Many times the fault rests not with the theory itself, but with a person's interpretation or usage of it.

 

[FDG]

any kind of slightly complicated game-theorethical model will give either a bayesian (i.e. probabilistic) or mixed-strategy result (i.e. a result which tells you that you're better off when you vary your strategy, with a given proportion between games played following strategy A and strategy B). Thus it does not deterministically predict what people will do in a given situation, rather what kind of reaction would give you an advantageous outcome if:
- such game is repeated N times (for each set of assumptions you can reach a satisfying N, varies accordingly to which game you're analyzing)
- every assumption holds

what I'm saying is...perhaps you have a wrong idea of what game theory actually is.

Also I think people should stop conflating game theory with the prisoner's dilemma. Its modern version is better represented by problems such as El Farol Bar or Signaling games

 

[xisnotx]

I'm just interested in how you came to this view about theories and theorising, why dont you think any of them are valid and if you dont think it is why is T one of your identified functions?

I believe it's the distinction between Ti and Te. All thought is a system. All systems are flawed. There is no such thing as "real reality", or if there is humans are incapable of thinking it. Science, math, etc are all flawed in one way..they are human constructs of what reality is. How can a human, or humans, know what real reality is, if they have never experienced it? Read the allegory of the cave, the idea is similar. btw, I'm not dismissing the systems of thought, I'm just aware of how they are limited. Not the same thing at all...

Do you consider a theory to be an artifice or invention which if it is consciously adopted and deployed by enough people is fact/reality rather than theory?

yes and no. If enough people believe in Jesus, then for all intents and purposed Jesus is real, even if he was fake. Jesus is just another "system of thought". Systems of thought, regardless of how illogical they are, are more influential to the world than logic is.

I would suggest you make an attempt to understand something before you pass judgment.

What do you think the purpose of the thread is? I have a basic understanding of what it is...what I want to know is what it "really" is..ie what people are doing with it, and if it's worth my time delving into it.

Theories are not simply conjectures or guesses about the nature of things, but tried and tested models which have solid foundations and have yet to be disproven.

To me, this sounds like "theories aren't guesses, they are guesses that haven't been disproven so are therefore considered true". Have you ever delved into the philosophy of the scientific method? Admittedly, I haven't (well, I have, but not really..I still need to do a lot of reading to do before I can say I've "delved into it")..but I think it would be an interesting topic. The beauty of science, I think, is that it is so flawed.

Also, MP is correct in his assertion that those who attempt to use a theory in ways which are outside its scope are simply "using it wrong."

One man's trash is another man's treasure.

I guess my real question was/is "Should I try to learn game theory? Or does the amount of effort it takes to learn it make it not worth it?" I'm lazy, so I'm asking people I think of as competent to analyze the validity of it. If enough people say it's worth it...then I'd probably have to set some time aside and take a serious look at it. Basically, I'm analyzing what people's analysis of game theory is.

But honestly? I rather just analyze the analysis. It's a lot more fun than just analyzing the system. The system of thought is too technical and arbitrary for me sometimes...so I get bored.