Can we actually predict how real people think in Strategic Games?
In this Networks class, we have been learning a lot about how game theory works and how it can be applied to the real world. However, is there an actual way to tie theory with the real thing? A lot of times, the probabilities and the outcomes predicted by the theory turns out to be rather accurate and correct. However, there are also many different cases and experiments in which our predicted outcomes don’t turn out to be correct in real life settings. This article talks about and explores a bunch of these different cases where theory and actual outcomes don’t line up and why the outcome is the way it is.
It is interesting how the biggest mistake that most people make while trying to use game theory in real life situations is the reasoning that everyone else knows what you know and will act similarly to what you do. As in the first experiment in the article, Minerva does what any game theory savvy person would do by basing her decision off of what everyone else is thinking and doing. On paper, this works very well; however, in real life it doesn’t because most of the people in the game with her either didn’t care enough to try and seriously and logically think it through or they didn’t follow through with the iterations nearly as well as Minerva did. In a sense, her assumption that these people were educated and carefully thinking through well known theories was her downfall. Even though she did what most well-educated and very mathematical people would consider a very smart choice, she actually ended up making a very poor decision compared to everyone else. The cause of this rift between reasoning and actual outcome is, simply put: most people aren’t as smart or as well educated in these areas as we think they are. Almost everybody takes the logical and simplest choice and sticks with it. Based on the variety and type of the population involved in the game, none to a rather small percentage venture to go further that one “step of thinking” (In a sense, it is like an iteration – logically applying a certain theory or assumption over and over again to obtain a certain final ‘logical’ choice”.) This underdeveloped thought process is usually what causes an expected outcome based off of theory to be very different than the actual outcome.
Also, another interesting phenomenon is how people decide to split certain amounts of money. As game theory would suggest, the person with the first and final offer has the most power and predictably end up with almost all of the money. However, when experiments and tests were done, there was a quite a lot of situations in which the person who was being offered the price rejected the offer when game theory would assume that they would take the offer because something is better that nothing right? The problem with the game theory in this situation is that game theory doesn’t really factor in human emotions, such as pride and fear. When something seems unfair to us, it is easy for us to either demand a fair option or make both people lose out, including ourselves. This pride mixed with the fear of getting nothing, adds up to where the common situation comes out to be that the person who offers first offers a rather fair deal, and where the person that gets to accept or decline usually only says yes to a rather fair/decent deal.
Things such as predicting how the variety of human thought processes and how human emotions effects outcome of a certain game are extremely difficult because since humans are rather diverse and fickle, it is hard to say they always act this one way or the other. Not to say that game theory can’t aid in making calculated and better decisions. However, it is true that knowing game theory does sometimes does not really help when playing a game.
Article: How Real People Think in Strategic Games (PDF)
-MITHui7