How Cooperation Can Be Achieved In Competitive Environments
To cooperate or not to cooperate, that is the question— and the answer from Robert Axelrod’s Evolution of Cooperation
Humankind’s evolution as a superior specie has a lot to do with our high intellect as well as our ability to cooperate. In spite of the primal instinct to hold our interests dear at the expense of others, the positives created by the interdependence and cooperation among our peers often led to the subordination of personal interest for the sake of societal gains. However, cooperation is not easy to sustain considering our egocentric and sometimes sneaky characters. People will always search for ways to maximize their payoffs — sometimes without realizing cooperation has always been the best option for maximization.
According to Hobbes, cooperation could not develop without a central authority, yet there is not a central authority for every aspect of life necessitating cooperation. Ever since, the question of whether a decentralised cooperation can emerge was challenged with a separate field of mathematics, game theory.
There are millions of scenarios where pursuit of self-interest leads to a poorer outcome for that person as well as for the others in comparison with cooperation. A famous representation of such scenarios is the Prisoner’s Dilemma game.
The game goes as follows: There are two prisoners who are interrogated separately but simultaneously. Interrogators offer both prisoners two options: confess their counterpart’s involvement in alleged crime or remain silent. If prisoners cooperate with one another and remain silent they will both receive 1 year of imprisonment. If one of them bails out and other remains silent, the snitch will be released whereas the silent prisoner will receive 20 years. If both of them bails out on one another, they will both receive 5 years. You, as Prisoner 1, are asked to decide.
Prioritising your self-interest without considering what happens to Prisoner 2, there are two possible actions your counterpart may take which you have to consider to develop your own strategy. If you believe that Prisoner 2 will remain silent, it seems better for you to confess his crime as you will be released whereas if you also remain silent, you will stay in prison for a year. Similarly, if you believe Prisoner 2 will bail on you, again it seems better to bail on him since you will receive only 5 years, whereas if you remain silent your punishment will be 20 years. So, no matter what Prisoner 2 does, you should confess his crimes, right? Not really. The flaw of this logic comes from the disregarding the logic of Prisoner 2, who would also think he should confess on your crimes no matter what you will do. As a result, you will both bail on one another and receive 5 years, whereas if you both cooperated and remained silent punishment would be only 1 year. In this case individual rationality leads to a worse outcome for both than is possible. Hence the dilemma.
How to determine the best strategy for multiple sequences of Prisoner Dilemma?
Prisoner Dilemmas, although not in literal sense, are everywhere in our daily and professional lives. In core, every scenario where there is lack of pre-settled trust and personal interests of two parties partially collide with one another has the same dilemma: A business relationship, or even a romantic affair can be subjected to it; however, in real-life there usually is not only one sequence of decision-making — there are many of them for the same example.
To put things into context, let’s give a more real-life like example to the Prisoner Dilemma. Assume that you and your housemate has decided to determine the payment of your monthly Netflix subscription with a game. Each month, both of you simultaneously put an envelope on the table, after which both envelopes will be opened. You may put inside the envelope a) the total monthly amount of the subscription fee b) nothing.
If both of you chose a, then each of you will receive the half of your money back and subscription fee will be paid. If one of you chose a and the other chose b, the subscription fee will be paid solely by the chooser of a and chooser of b will enjoy a month of Netflix for free. Lastly, if both of you chose b, subscription fee will not be paid and you will be doomed with the lack of Netflix & Chills for an entire month. Same logic as Prisoner Dilemma, however as you can realise this game will take place every month, and over the course of months you will get to learn if your housemate is a horrible person who never pays or someone with deep pockets who is happy to sponsor your Netflix & Chill sessions — and vice versa.
Determining a “best strategy” requires the mention of some ground rules. First off, it cannot be considered separately from the other player’s strategy. Second, the pessimistical assumption of the other player making the feared move will lead to you making the same move as well, causing mutual punishment. A chess player can safely use this assumption as it provides the ability to foresee your opponent’s move in a game like chess, where players’ interests are in total conflict — there cannot be two winners or two losers. On the contrary, situations represented by the Prisoner’s Dilemma game have personal interests that are not in total conflict as mutual cooperation is rewarding.
The winningest strategy: TIT FOR TAT
To find out the best strategy, a multiple sequence Prisoner Dilemma tournament was organised with participants submitting their favorite strategy, and the winner was surprisingly the simplest of all strategies submitted: TIT FOR TAT, the strategy which cooperates on the first move and then mirrors the other player’s previous move.
There are four main attributes of TIT FOR TAT behind its success:
• “avoidance of unnecessary conflict by cooperating as long as the other player does,
• provocability in the face of an uncalled for defection by the other,
• forgiveness after responding to a provocation,
• clarity of behavior so that the other player can adapt to your pattern of action.”
The unique combination of being nice, retaliatory, forgiving, and clear balances the weaknesses of each attribute: Niceness prevents unnecessary trouble while retaliation discourages the other side from resisting not to cooperate. Forgiveness allows to restore mutual cooperation while the clarity offers intelligibility and transparency to the other player, thereby eliciting long-term cooperation.
Some life lessons
21st century people, while functioning as a society under a social contract, has embraced the ideals of individualism and self-determination of one’s true potential. We tend to think that if everyone pursues their own interest, this competition will lead to the most efficient and collectively-rewarding result — this is how Adam Smith envisioned it back in the day. Yet, thanks to the game theory we now have discovered a third way: cooperative competition (or competitive cooperation). This offers us to get the best result for ourselves as well as for the others, and unlike the sole pursuance of self-interest, with more people acting in the same manner this strategy’s personal outcome can only get better.
Levent Aslan, Social Media Guy
