Applying Game Theory into Everyday Life

Since game-theoretic decision making models are my primary focus on my readings now, I will attempt to incorporate it into the whole gender-theory and gender-relations thing I’ve got going on here on this blog.

Now the whole gist of my ideas has been gender equality. Basically, I will interpret an extended form of decision making through utilities and preferences in dissatisfaction within two people in a relationship, named “Player A” and “Player B”. Examine Graph 1.
Game-theory chart
Graph 1. Extended Decision Making Model. (Brackets, denote utility preferences. First number for player A, second number for player B).)

Lets say within a relationship, there is some dissatisfaction from a certain partner in the relationship. Lets assume that initially Player A has two choices, to cooperate (and thereby maintaining the status quo) and to defect (to confront the other player in emotional blackmail). If Player A chooses to cooperate and maintain the status quo, lets assume that Player B is pretty comfortable there and is the best state that Player B wishes to be in.

If Player A defects, then Player B has to make a response, either to cooperate or to defect. Cooperating means that Player B capitulates to the demands of Player A thus apologizing. On the other hand, to choose to “defect” would mean that it would lead to a fight, a break-up, or the equivalent of that.

Now we also assume because Player A is dissatisfied with the status quo (such as Player B is not calling often enough, Player B is seeing other people), Player A prefers that Player B changes i.e. Make-up & Apologize. Thus, Player A’s utilities are such that Make Up > Status Quo. However, Player A does not want to break up with Player B, thus Status Quo > Fight.

My argument is that in the case of emotional blackmail, all Player B’s should make it clear to Player A that Player B’s utility at “Make up & Apologize” is less than the utility at “Break-up”. i.e. x < y, OR in simple terms, “if you want to leave me then go ahead”.
Player B apparent utility: Status Quo > Break-up > Make-up & Apologize

Its really a very simple reason. If Player A knows that Player B will definitely prefer “Break-up” over “Make-up”, then Player A will choose to cooperate as Player A prefers “Status Quo” over “Break-up”. However, if Player A knows that Player B prefers “Make-up” over “Break-up”, then Player A will definitely defect so that the state of “Make-up & Apologize” is achieved, rather than “Status Quo”. If Player A knows this to be true over many issue areas, Player A can force Player B to “Make-up & Apologize” for as many issues as Player A knows that Player B will cooperate. This is an undue power imbalance within a relationship, and leaves Player A at a disadvantage.

Now, the shortcomings of this reasoning and game-theoretic modelling. Just because you know that taking a hard position such that making known that one prefers defecting over cooperating at Node 2 does NOT mean you should ALWAYS prefer defecting over cooperating. If you should abuse your position within a bilateral relationship such that the status quo becomes unbearable to Player A, the utility of Player A might change to the following:
Player A: Make-up & Apologize > Break-up > Status Quo

In this case, anything is better than the status quo to Player A. If Player B true utilities are such that Break-up is undesirable, i.e.
Player B: Status Quo > Make-up & Apologize > Break-up

then Player B ought to act so that Player A’s utilities does not fall change that drastically. Pragmatically, that means that Player B ought to cooperate in certain cases and conceed during some contestations, and stay hard in pivotal cases, especially during emotional blackmail.

One last note: Player A and Player B are not necessarily Girl and Boy respectively. It works both ways.

I really have too much spare time.



  1. Why would a girl prefer to defect and the have the boy makeup? (payoff of 3 with makeup)
    Without that, the dominant strategy would be to cooperate. Maybe that’s why some guys hate girls who whine to get attention, because it makes for an unstable relationship.

    Maybe you could model it as a multi-stage (to infinity) game theoretic model. The payoffs for patching up could depend negatively on the number of times the girl has defected. That is both intuitive, agrees with common sense, and can also model long-term abuse.

  2. You’re right, I haven’t given reasons for why I assume “Breakup” to be preferable to “status quo” for Player A (girlfriend).

    It’s essentially emotional blackmail. Player A (GF) is trying to get Player B (BF) to feel bad and getting BF to capitulate to GF’s implicit request. Thus, when BF “cooperates”, he’s giving in to her and she has her way.

    Extending it to model longer-term decisions is an excellent idea! I was thinking that it just reproduces itself on the same payoff scheme after BF cooperates.

  3. There is no reason why the payoff to Player A from Break Up should be lower than the payoff from Status Quo. Because in reality, one reason people do choose to break up is the possibility of someone better out there. If that is the case, then payoff from Break Up should be:

    U(BU,A)=1+3*P(n), where U(BU,A) is payoff to Player A from Break Up and P(n) is the probability of finding someone new. The new function incorporates the expected payoff from changing partners, which is the product of the probability of finding someone new and the payoff from Apologize (assuming that Apologize represents the best possible payoff in a relationship, though it could be higher which would make the following conclusion stronger). Summed with the payoff/unpleasantness of breaking up, the new payoff may well exceed the payoff from Apologize.

    What this means is that Player A’s strategy must take into account his(her) belief of his chances of finding someone new. If Player A holds a belief m that he has a probability P(m) that he will find someone new and 3*P(m)>1, then the perfect Bayesian Nash outcome should be that Player A will choose to Break Up.

    In the end, a strong bet is that Bayesian equilibria have to get involved somewhere. And once you start along this road, you’ll have to accept that Player B would also be thinking of finding someone new. Add the uncertainty of the value of the payoff from finding someone new… Have fun.

  4. Correction: Summed with the payoff/unpleasantness of breaking up, the new payoff may well exceed the “payoff from Apologize”.

    It should be the payoff from Status Quo.

  5. Interesting. Shows how relationships exist as a construct within which both ‘players’ are playing a game of manipulating each other to get what they want, making a decision based on how one player can best serve their own self interest using what knowledge they have of the other’s interests

    thus, the people within the relationships are not standing equal with each other, not supporting each other — but rather just playing a game of manipulation in self interest

    and children are born and raised and educated within such an environment — being taught to manipulate and deceive in self interest rather than evaluate themselves, others and their world within common sense and self honesty and to then direct themselves within what is seen to be the most practical action to take within what is what is best for all in equality and common sense

  6. this is interesting!

  7. Thanks for youг mawгvelous posting! I actually enjoyed rezding it, yyou are a great author.
    I will always bookmark your blοg and mayy come back iin the future.
    I want tto encourage one to сontinue your gгeat job, have a nicе evenіng!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: