Decision Making: Is It All 'Me, Me, Me'?

According to classical game theory, decision makers invariably act in their individual self-interest, leading to "Nash equilibrium". However, as is explained in this article, "psychologists at the Universities of Leicester and Exeter have recently found evidence that this assumption is not necessarily true." The lead researcher said, "We have shown that, in some circumstances, decision makers cooperate in their collective interests rather than following the purely selfish predictions of orthodox game theory."
This seems a reassuring message, telling us that we are not necessarily the selfish beings as the model of classical game theory predicts. Or are we?

I feel like this non-self driven action away from the predictions of game theory, besides some rare cases, is only observed due to the use of over simplified games. If a bi matrix predicts the equilibrium of a game to show in some way that a player will not give money to a specific group it most likely does not take into account factors such as guilt, self satisfaction, and other personally motivating things that though are not directly in the interest of the self as far as the simple give or don't game goes, in a less simplified version of a given game the actual equilibrium may show the result observed.

I agree with the article in thinking that, under some circumstances, humans do not always act selfishly. Psychologists have done numerous studies to suggest that sometimes animals act for the good of their family members or for the herd or pack. Even in a basic, introductory Psych course, we learn that there are instances of "social facilitation" where individual goals take a proverbial back seat to the group's goals.

However, although I think that this does occur for sure, it is definitely not a phenomenon that takes away from game theory. On most occasions (over 99% of all, I'm sure), our selfishness takes priority and game theory will hold. To look for examples of selfishness, take into account every game we play in class as proof. So far, the "ideal" situation has never been reached because almost everyone decides to take the "selfish route."

So, game theory will stick the vast majority of the time because most instances of the group overriding the individual's goals are few and far between.

Maybe being selfish can be in the best interests of the group? For instance, there are things that are in everyone's best interest... so it makes perfect sense that we could all collaborate to make things better for everyone. But that seems too obvious. .... Have you ever thought that maybe people looking out for their own self interests might also make things better for everyone? An example Im thinking of is along the lines of companies competing to make the best product. If each company just wants to make a profit for itself and beat out its competitors, that is purely selfish. However.... each company is going to then have the motivation to build the best, most appealing product for their customers. Because the customers want the most efficient and best product for the best price, that is what the companies will create for them.
The society is able to benefit from these products that were only developed to be the best because of the companies selfish competition.
So what's wrong with a little selfish competition?

The article is intriguingly vague in that it doesn't specify which games were shown to have a greater weight with the Team Reasoning theory over the Classic Game Theory. I think it could be possible that in certain situations, acting in one's own "self-interest" may very well happen to coincide with the best interest of the group, and thus the illusion of the Team Reasoning effect emerges. I suspect the issue largely revolves around the consideration of what constitutes an individual's own self-interest -- does it only consider strictly the direct and isolated outcome of the game itself, or does it also account for situations outside and beyond the outcome of the immediate game situation?

Consider as an example the situation of a company struggling somewhat in the current economic situation. There is an employee A who works at the company in a cubicle, and a manager B who is responsible for A amongst other workers. A also has a family and loved ones. Further, A dislikes his job.

The situation arises in which A is asked by B to come in to work on Saturday to continue working on a very large project which is very important to the company.

In the immediate game situation on the finest level of detail, A has two options -- go to work on Saturday (a1), or not go to work on Saturday (a2). If acting only in his immediate best interest based on the isolated circumstances of the game, A will clearly choose a2, as this optimizes his gain from the situation (he is most happy by not going to his hated job on Saturday). However, in reality, A will (usually) consider more than just the immediate situation. If he chooses a2, his immediate payoff is negative -- the choice is effectively an investment in the future. By choosing a2, A hopes that at some point down the line, he will receive a promotion, or a pay raise, or a bonus, which he can use to help his family.

Upon cursory inspection, it appears that the choice of a2 is a relatively selfless action in an effort to better the situation of his "Team." However, the personal (monetary, social, emotional, psychological...) benefit in having a stable and provided-for family, a better job, and/or increased wages which A hopes to gain by choosing to act in the a2 manner make it clear that while his choice seems to be for the sake of the "Team" and not for personal gain, A has still ultimately made a choice in an effort to maximize his own personal payoff.

Brad, this is a good example, but I think through your "cursory inspection" you've actually shown that game theory still holds. The error was in your first assignment of the payoff matrix. You calculated the payoff matrix based on the immediate payoff, but then proceeded to determine the outcome of the game by taking into account both the immediate and delayed payoff. This is what leads you to determine (in quite a roundabout way) that A is still making a choice that benefits him personally. If the initial payoff matrix was accurately created to take into account all forms of payoff, "(monetary, social, emotional, psychological...)", the 'game' would still make sense according to classic game theory. This game is not a case of action for the good of the group at the hands of a diminished payoff for the individual, rather, it is a case of action for the maximum benefit to the individual that results in a good payoff for the group too.