The Cooperation Game

February 2018
Ivan Obolensky

Evolution is often told as a story of fierce competition for limited resources, resulting in natural selection that favors those individuals most able to take advantage of the environments that surround them. Only the fittest survive.

Although this is one way to look at evolution, in reality, evolution is much more a story of cooperation than competition. This article will elaborate on this point of view.

Why “survival of the fittest” became the catchphrase for evolution can be found in the extant thinking of the 19^th and early 20^th centuries. During that period, the individual was the primary source of economic power. The prevailing paradigm was that “Man, the individual, was the measure of all things”.

Individual decisions were thought to be the result of informed and rational self-interest. This understanding also underpinned many theories in economics and the social sciences.

For instance, the Efficient Market Hypothesis has its roots in 1900 France. It states that asset prices fully reflect all available information, that those who trade and invest are rational in their choices, and that people act exclusively in their own best interests.

Tests on large numbers of individuals should have reflected the predominance of self-interest in decision-making, only the results told a different story.

Self-interest was present in many choices that people made, but individuals routinely took into account other factors, such as fairness, or the impact of their decisions on others.

This was a surprising result, given the prevailing understanding, but not altogether unexpected if one simply observes how most of us decide in our own minds. We take others into account in many of our decisions.

Evolution, too, points in the direction of cooperative activity. Multicellular organisms could not have evolved if the best survival strategy was strictly to look after oneself.¹

In most animal groups, including primates, competition is the norm. Humans are different in this regard.

A comparison study of two-year-olds and chimpanzees and orangutans had each group perform sixteen tasks in two categories. The first measured understanding of the physical world, and the second, an understanding of the social world. The social tests concerned imitation, communication, and the ability to read the intentions of others. Human children were not overall the more intelligent but excelled specifically in social intelligence, a skillset that allowed them to cooperate. Because the study was performed with two-year-olds rather than older children, it would seem that cooperation is not simply a learned behavior, but the result of genetic predisposition. Humans have a built-in bias to collaborate with each other.²

As our understandings of the world change, what we emphasize and focus on changes as well.

Game Theory developed into a unique field through the work of John von Neumann and others, particularly during the 1950s. It quickly expanded from competitive non-cooperative games to cooperative games, evolutionary games, and network games. It uses mathematical models of conflict and cooperation to understand logical decision-making in humans, animals, and computers.

Although many of the problems that Game Theory addresses are theoretical, some reflect real-world challenges. The social dilemma is one that not only describes and encapsulates the frenetic gyrations of current society, but whose solutions will determine and influence the future of our civilization more than any other.

A social dilemma is a situation where the individual profits by being selfish, but if that selfishness pervades every member of a group, than the group overall, as well as the individuals in it, also suffers.

This issue can be expressed using Game Theory in two concepts.

The social payoff for the individual is higher for uncooperative behavior than it is for cooperative behavior, regardless of what another does.
The overall payoff to the individual is less if everyone fails to cooperate, than if everyone did.

Example:

Joe steals. Joe survives well on his loot.
Everyone steals. The quality of life for everyone is much less overall, including Joe, who is robbed all the time as well.³

Social dilemmas are of major importance because they reflect the core issue of the individual vs. the group, and what happens when self-interest is in conflict with group interests.

Although the dilemma might seem easy to solve, there is more to the problem than simply dictating that the needs of the many should supersede and take precedence over the needs of the individual.

What happens to the individual when the individual’s needs are overlooked? With the waning power of the individual and the corresponding rise in the power of the network, how far will individual rights, freedoms, and pursuits be erased?

Taken to extremes, all the individual will do is labor, or in the event that sophisticated computers do everything, humans would simply hang about until the cost benefit analysis shows that having humans around is an expensive luxury. What do we do then? More specifically, what will you do then?

Another element is that humans do not do well in prisons. Why do we imprison criminals other than to punish them? And yet, as our freedoms erode, and walls are built around us in the form of what we must and mustn’t do in every category of action, are we not creating a prison of our own making? Mentally, psychologically, and physically, it is shown to be unhealthy. (Note the section about the dark secret of early pioneer life in the New World in “The Motivation of Genius, Part II”.) What happens to us all when there is nowhere to run, nowhere to hide? Self-interest is not merely selfish. With no sense of autonomy and personal stake in an outcome, there is no freedom to choose and we find ourselves ill, purposeless, and depressed. We die.

The eradication of the individual’s power of choice and acceptance of whatever society dictates lies at the heart of dystopian visions of the future.

How about the reverse? In Game Theory, those who decide to be selfish are called defectors.

One of the concepts that has permeated literature and philosophy is that life is a jungle. Man, if left to his own devices, will devolve to the strongest individuals gathering up all the resources for themselves. This idea has been used as a justification for a strong government because only through strong government, a monarchy, or at least a third party who can apply sufficient force, can the baser instincts of the individual be controlled and civilization flourish.

Let us take the worst case of the above: a western town is invaded by outlaws, the ultimate defectors. The usual story is that a gang of gunmen shoots up the locals and takes over. There are two ways this story develops. Either a lone sheriff takes them all on, and order is restored, or everyone is massacred. But is this the way real life works?

Not necessarily. The outlaws need food. They need bullets, weapons, and liquor in the saloon. Where does that come from, and where will it come from in the future? Once all the resources are used up, what will happen to the gang? They will either starve, or move on. What is missing from many of these stories is that life continues, and that infrastructure must exist to sustain their own lives going forward. This we might call the context of the game.

In such scenarios using computer simulations repeated many times, the defector-cooperator dynamic is cyclical.

For example, individuals in a fishing village cooperate by fishing, and the catch is shared with the whole village. Suppose one villager (a defector) decides not to fish. He still eats, only he doesn’t have to do any work. Suppose his friend decides this is a workable strategy, and he decides to not fish as well. As more individuals defect rather than cooperate, a tipping point is reached when there is not enough fish to feed everyone, and the village starves. A meeting is called. The defectors are either thrown out, or forced to cooperate. The cooperate strategy reasserts itself as the predominant mode of conduct.

Climate change, immigration, and homelessness are examples of today’s social dilemmas. The solutions that are often put forward require more regulation by governments to solve them. Although this is a possible solution, it is still grounded in the thinking that humankind, if left to its own devices will devolve into a dog-eat-dog scenario, similar to the outlaw one mentioned above. That may not be the case.

Game Theory uses various examples to demonstrate the power of choices. One that is similar to the Social Dilemma is called The Prisoner’s Dilemma. A pair of criminals are put in separate cells and told that if they turn in their partner, they will be set free, while the partner will get 10 years. If, as a result, both turn each other in, then they’ll both get five years, but if neither turns in the other, then each gets only six months.

The dilemma is that it pays for both to defect to avoid the ten years, only the result is that they are worse off than if they had both cooperated with each other and held silent. This is similar to the outcome of social dilemmas.

In a single instance, it might be best to defect, but what if this type of choice has to be made many times?

In the 1980s, computer simulations were used to pit various strategies against each other. A strategy was considered more successful if it produced less prison time overall when compared to others. For instance, one could have a strategy of continuous defection, or another of always cooperating. The most successful strategy was Tit for Tat. The first meeting has the default setting of cooperation to be followed by doing whatever the opponent did the next time. For example, if the opposing strategy defected, then so would Tit for Tat, but if it cooperated, Tit for Tat would do the same.

Simulations run over many iterations showed that if everyone had a defection strategy, which represents the ultimate in self-interest and competition, and a Tit for Tat (retaliation) strategy is introduced, the Tit for Tat strategy defeated all others time after time. Simulations also showed that small clusters of cooperators that formed did better in terms of survival than defection strategies, and that the clusters created more cooperators, who also prospered.

Further, using the Tit for Tat strategy as a base, a mutation strategy was developed called “Generous Tit for Tat”. This strategy incorporated a probability algorithm that determined whether the entity would retaliate, or continue to cooperate in spite of being defected against. Because the decision was random, opposing players didn’t know what it would do and tended to cooperate.

This strategy turned out to be the most optimal and is to some degree a reflection of the real world. We never know who we will meet, and cooperating initially is usually the better play. Forgiving some defections, such as retaliating nine out of ten times, led to the best outcome over the long term.⁴

Another key concept that has come out of this research is that of reputation and interconnectivity.

Humans like to connect, and connection introduces some measure of trust, particularly if many transactions are executed successfully over a period of time. This establishes a reputation, and when people decide whether to trust or not, reputation can be decisive in terms of cooperation.

Connectivity via social media and other Internet-based platforms creates a sense of proximity to the social dilemma issues of today. Connection internalizes the issues, and people take them personally as a result. It is this sense of closeness to the issues that is the source of the extraordinary volume of upset we see in the world today. Interconnectivity promotes opinion and subsequently engenders doing something about it.

In spite of many indications to the contrary, this may turn out to be a positive. Cooperation strategies win over defection strategies and by connecting with the issues, the issues become ours to solve.

The question becomes “what do we value as individuals and as members of society?” Also where is the balance between the individual and the issues confronting society?

In addition, social dilemmas are also not simply about social ills.

Cryptocurrencies are an example of the social dilemma in regards to transacting business without fear of being ripped off and without the need for a third party to intervene or verify a transaction.

Cryptocurrencies utilize Blockchain technology. A Blockchain is essentially a giant secure database that is shareable. It allows users to record information without fear that the information will be altered or corrupted.

How Blockchain technology will aid in social dilemmas, obviating the need for regulation is an interesting concept that has potential. It is a new way to establish trust, not as the result of centralized institutional verifications that are open to corruption or dishonesty, but rather through incorruptible protocols, cryptography, and computer code. It is particularly effective when mass collaborations are required.⁵

Perhaps the most important potential for this technology, in my opinion, is in data preservation. What was written, recorded, or videoed can be proved to be valid using this technology. Tampering with digitally preserved information is a danger that has become more potent as the world stores information digitally with no way to know if that information has been altered, destroyed, or partially withheld. Imagine being certain that your copy of 1984 is the same as the original print edition or that the video presented as evidence in court was not altered by a vested interest to change the outcome? If 99% of all money is electronic, as well as 66% of the economy, what prevents your money from simply disappearing, or being transferred into someone else’s account?⁶ How do you prove you had it in the first place? Blockchain technology can prove it, without a bank, a government, or anyone else verifying or contradicting it. It will allow cooperation on a whole new level.

In summary, cooperation is not just a successful strategy. The individual is also aided by adopting it. Cooperators create better opportunities for themselves than non-cooperators. They are preferred as partners and as leaders. But when is cooperation too much?

The benefits of cooperation vs. self-interest are in itself a cost-benefit analysis. At some point a balance is achieved where the benefits of cooperation no longer outweigh the cost to the individual in terms of well-being. Where this point is exactly is unknown, and given the human tendency to overshoot in one direction, or the other, means adjustments will need to be made going forward.

Wealth disparity and many of the issues of today are social dilemmas in various guises. How will we solve them? Blockchain technology may be an unlooked-for answer.

Ultimately, humans have an innate ability to cooperate. What allows us to live together in such numbers is our social intelligence. For myself, I have confidence that this capability will prove decisive going forward.

It will all work out eventually because that is what we as humans do. We work it out. We were born to.

N.A. (2017), Game Theory: An Overview, A Complexity Labs Publication. Retrieved January 19, 2018 from http://complexitylabs.io/product/game-theory-introduction/.
Haun, D., Rekers, Y., Tomasello, M. (2014), “Children Conform to the Behavior of Peers; Other Great Apes Stick With What They Know”, Sage Journals. Retrieved January 19, 2018 from http://journals.sagepub.com/doi/pdf/10.1177/0956797614553235.
Game Theory cit.
Kay, R. (2011), Generous Tit For Tat: A Winning Strategy, Forbes. Retrieved January 19, 2018 from https://www.forbes.com/sites/rogerkay/2011/12/19/generous-tit-for-tat-a-winning-strategy/#5a30e40b66eb.
Colchester, J. (2018), Blockchain: an Overview, A Complexity Labs Publication. Retrieved January 19, 2018, from http://complexitylabs.io/blockchain-book/.
Armstrong, M. (2018). “66% of the Economy is Already Electronic & 99% of Money is Electronic.” Armstrong Economics. Retrieved January 19, 2018 from https://www.armstrongeconomics.com/world-news/cryptocurrency/66-of-the-economy-is-already-electronic-99-of-money-is-electronic/