Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma
nature.com
nature.com
The main novelty about this work is that we had 100 people play repeated PD for a month instead of <1 hour which is typical in behavioral experiments, in an attempt to make rather stylized cooperation experiments in the lab closer to real life (admittedly, it is still only a small step.)
The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."
You say as a summary :
> a minority of nice people can make everyone better off
Was it possible for you to identify the "nice people" group vs the other group?
If yes, overall which group was better off ?
Once the non nice people start cooperating, i guess you can't make the difference between a nice person and a non nice one
We can tell the difference between the two types of people, because the nice people never defect first.
Finally, with regard to identifying these types in general, one could use the (very large) dataset from this experiment to train a simulated agent that plays against a human, and quickly determine which type they fall into.
In other words, over time, the "compound interest" on the gains the nice individuals lose gets rewarded to the group. Any individual in the group who can game "niceness" benefits the group at the expense of the nice individual. "Good guys finish last," or something of that nature?
Stabilized long-running cooperation seems like a very bad desired(?) outcome, by the way; is that what you're saying is a good thing? Pareto improvements are usually realized / discovered / gained by individuals who eschew tradition and playing nice.
However, it's not true if you look at it from a long-term perspective. If the nice players know that not being nice would cause everyone to converge to the highly inefficient Nash equilibrium, then they are actually better off by being nice (and promoting cooperation) than they would be in the latter case. So even though they're not doing as well as the "selfish" players, they're better off than they would be in the dystopian alternate reality. (Ahem, ahem.)
I'm thinking about same game setting, except that players are asked to give their mugshot and real or fake mugshots of males, females, people different ethnicity, looks and ages are shown to them.
In our case, however, we made the same people play for a month. I believe that learning effects strongly dominate priming over that time period.
Give players the ability to remember players they have engaged with and add a third option: 1. Cooperate 2. Defect 3. Wait (defer, don't engage, don't play). New players will seek out players and offer to engage with them. When you defect, your reputation history declines, and when you die, your reputation history resets to zero and you lose the option to defer.
HYPOTHESIS: Cooperators will wait/defer, Defectors will wait/defer, new players will choose to cooperate or defect, and since defecting against defer is an all-but-certain lose, new players will choose to cooperate hoping the other player will too. Cooperators will naturally organize into groups of cooperators and flourish. Players in cooperator groups will cooperate tit-for-tat with known cooperators and will wait/defer with known defectors or new/unknown players no one in the cooperator group has engaged with (call it the no-fools strategy). New players and players who have been struggling in the wilderness outside of cooperative groups will want to get into a cooperative group and so they will seek to engage with players using the tit-for-tat strategy until they find a cooperative group (call it the pay-it-forward strategy). Fools who don't learn will stumble around in the wilderness until they eventually die off.
First, we discovered in the experiment that PD actually has 3 actions. Cooperate, defect, and ... wait as long as possible to defect to punish your opponent who is not cooperating. This happened because participants are playing in a real-time web app, and they have some time to make a decision. Although there is a lot of literature on how costly punishment can be used to enforce cooperation, it was pretty amazing how it emerged organically in our study after a few days. Someone could certainly grab our dataset and get a free publication out of just that.
Also, given the month-long design of the study, we were thinking of something where players get to stay on or "die" based on their daily score, and therefore the population evolves based on who gets the highest payoffs. Evolutionary game theory has some ideas about this, but they are all theoretical or simulation, and not conducted with real people. I'm actually not sure if cooperation would be sustained or not in this design, but I agree that it would go a long way toward pinning down the origins of cooperation. I also agree with you that groups or network structure would affect the result a lot, allowing cooperative bands to flourish even when defectors are running around.
https://en.wikipedia.org/wiki/Jonathan_Haidt
NB: Also see this timely post on "information theory and the foundations of life" https://news.ycombinator.com/item?id=13496133
http://www.ted.com/talks/jonathan_haidt_humanity_s_stairway_...
Also this Edge article on "Contingent Superorganism" https://www.edge.org/response-detail/10386
To add the option to "wait/defer", you could include informational incentives/variance where players gain the ability to increase their perspective and degrees of freedom by being privy to more information sooner. For example, once a player achieves some level of points/reputation, they gain the ability to see the other player's move before they move -- this would in effect be giving them the power to defer. The other player may or may not know the extent of the other player's abilities at that time -- one or both may have the ability to see (defer), and neither may know for sure. For example, rather than explicitly stating upfront to new players that through their play they can improve their ability to see, let players discover this once they cross a threshold. Further increased powers might include the ability to see into the other player's past games, see their group's dynamics or portions of the wider network, and communicate with other group members and/or the other player during games.
As stated by Axelrod in EoC, one feature of the tit-for-tat strategy is that it's easy enough that it can be understood by anyone, and it's simple enough that it can be communicated/signaled from one player to the other through their play. It would be interesting to see how learned players -- who have gained the ability to see -- use or abuse their elevated powers of sight, either by working to better signal/communicate their intent and teach the winning strategy to the less enlightened players or by choosing to try and exploit the naivety of the other player who may be relatively blind.
The 3rd option used most widely in business today. Defectors are "punished" by refusing to do business with them after a defection.
Also, how often is the number of rounds known in studies like this? From an algorithm perspective of maximization of points for computer agents, typically the number of rounds is random.
So the cooperators' strategy looks like tit-for-tat, but I don't think it really is because tit for tat means if someone started cooperating again after defecting, they would switch. We think it's actually a strategy where if you get defected on, you just stop cooperating altogether in the future (at least for this length of a game). However, because we never see people switch back to cooperating after defecting (after the first day where people are trying random things), we can't distinguish between these two strategies. But effectively it doesn't really matter.
A game with a finite number of rounds is a finitely repeated PD, for which the NE is to all defect. An undetermined number of rounds is ~= an infinitely repeated PD with a discount factor, for which there are many ways to sustain cooperation according to the theory.
1. Are there any plans to make the dataset available?
2. I missed this, what incentive is there for players to "unravel", or move the first round of defection earlier? My naive thought is that turkers want to maximize monetary payoff by getting a high bonus (1 cent for each 2 points earned) and playing quickly (to free up time for other HITs). From your payoff matrix, both players cooperating all ten rounds in one game each get 50 points. A player could instead defect on round 10 to get 52 points. He could also defect on round 9 and expect mutual defection on round 10 to get 50 points. But if he defects even earlier on round 8 and expects mutual defection afterwards, he gets 48 points. In this experiment, earlier and earlier defections seem worse than cooperating on all rounds, even for the defector.
If your opponent is going to defect at some point anyway, then if you defect right before he does you will be better off (at least for that game). That's the pressure to unravel.
Suppose both you and a partner get arrested whilst trying to rob a bank. Without a confession from one of you, the police can only convict you of a minor offense (say trespassing).
The police place you in different cells, and offer you both the following bargain. If you confess to the robbery, and your partner does not, you get a complete pardon, and your partner gets a full sentence (say 120 months in jail). If you both confess, you get a slightly reduced 100 months in jail. Finally, if you both remain silent then each gets a 3 month sentence for trespassing.
You both now have this tension between individual and collectively optimal action. Your personal sentence is smaller (independent of what your partner does) after confessing, but you both benefit from coordinating to not confess. You can draw the payoffs in a matrix (left is row player and right is column player) to visualise this.
c c'
c 100,100 0,120
c' 120, 0 3,3
Even if there is a repeated interaction for a finite amount of time, you can expect rational self-interested agents to be unable to collaborate in the final period. This means they have no way of punishing in the penultimate round, and so by induction you would expect 'unravelling' to non-cooperation right from the beginning.
This paper found that a subset of the population appearing not to be self-interested was sufficient to prevent this complete unravelling. Even a selfish player may wish to collaborate in early interactions if there is a high enough chance that they won't be immediately punished.
As an adult I understand that this isn't the point of the story, and the police-and-prisoners aspect of the story is completely extraneous to the point trying to be made. Still, I can't help but think that the police-and-prisoners is a bad example of the broader point, since there's always a third set of interests (that of the police, and of society in general) which is callously (and immorally) tossed aside in the phrasing of the problem.
Here's a fascinating quote from a book about Kenyan graft [1]:
--------------------
Where does each individual draw the limits of his or her compassion, beyond which duties of kindness, generosity and personal obligation no longer apply? I was raised in a household where my parents drew them in totally different places, according to their very different characters and backgrounds.
As an Italian, my mother grew up in a country whose government had given birth to Fascism, formed a discreditable pact with Hitler, and launched itself on a series of unnecessary wars which left Italy occupied and battle-scarred. There then followed a seemingly endless series of short-lived, sleaze-ridden administrations. The experience left her utterly cynical about officialdom. Although she dutifully voted in every election, the malevolence of the system was taken for granted, and she would happily have lied and cheated in any encounter with the state had she believed she could get away with it. But no one worked harder for her fellow man, for in the place of the state she maintained her own support network. An instinctive practitioner of what sociologists call 'the economics of affection', my mother had a circle of compassion drawn to include a collection of lonely acquaintances. She visited their council flats bearing cakes, sent amusing press cuttings to their prison cells, queued at the gates of their psychiatric hospitals. Hers was a world of one-on-one interactions, in which obligations, duties, morality itself, took strictly personal form, and were no less onerous for it. The glow she radiated was life-enhancing, but its light only stretched so far, and beyond lay utter darkness. Protecting one's own was vital, for life had taught her that the world outside would show no mercy. She was not alone in her ability to get things done without the state's involvement. 'Il mio sistema' Italians call it: 'my system'. Italy is, after all, the birthplace of the Mafia, the ultimate of personal 'sistema', and my mother's mindset was instinctively mafioso.
My father, in contract, was typical of a certain sort of law-abiding, diffident Englishman for whom a set of impartial, lucid rules represented civilisation at its most advanced. He was raided in a country which pluckily held out against the Germans during the Second World War and then set up the National Health Service in which he spent his career, and his trust in the essential decency of his duly elected representatives was so profound that he was shocked to the core by British perfidy during the Suez crisis, and believed Tony Blair when he said there were weapons of mass destruction in Iraq. When, as an eleven-year-old schoolgirl, I mentioned - with a certain pride - that I usually managed to get home without paying my bus fare, he explained disapprovingly that if everyone behaved that way, London Transport would grind to a halt. Remove the civic ethos, and anarchy descended. A logical man, he saw this as the only practical way of running a complex society. It also, conveniently for an Englishman awkward with personal intimacy, enabled him to engage with his fellow man at a completely impersonal level. Not for him my mother's instinctive charm, the immediate eye contact, the hand on arm. He felt no obligation to provide for nieces and nephews, and had a cousin come up for a job before one of the many appointment boards on which he sat, he would have immediately excused himself. Nothing could be more repugnant to him than asking a friend to bend the rules as a personal favour. What need was there for a rival, alternative sistema, if the existing arrangement of rights and duties already delivered?
My father's world view was typically northern European. My mother's characteristically Mediterranean approach would have made perfect sense to any Kenyan. In an 'us-against-the-rest' universe, the put-upon pine to belong to a form of Masonic lodge whose advantages are labelled 'Members Only'. In the industrialised world, that 'us' is usually defined by class, religion, or profession. In Kenya, it was inevitably defined by tribe
[1] https://www.amazon.com/Its-Our-Turn-Eat-Whistle-Blower/dp/00...
The prisoner dilemma also ties into the tragedy of the commons: when multiple players decide to defect and exploit the commons to the fullest, then commons lose their value for all.
It is also related to temporal discounting - we prefer to pollute today and ignore the price tag we will face in the future.
We know so much about game theory yet we can't apply it in politics, because it requires long-term thinking and optimizing for the success of everyone as opposed to just a few. And we're in a point where words don't mean as much, we're flooded with propaganda from all sides and can't speak reason over the cacophony.
Another meta-conclusion I would draw from our paper is that current game theory is insufficient to explain what we observe in real life, including politics, negotiation, and so on. It is folly to apply hyperrationality (strict economic modeling) to real human behavior, and part of the goal of our work is to stimulate more realistic models of what people do in these situations.
Of course, many political situations also exist where companies donate heavily and then ask for such penalties to be removed so they can sell worthless securities or not have their emissions monitored anymore. Democracy turns out to be pretty thoroughly corruptible, (but so do other forms of government.)
So, people aren't strictly rational actors, and altruism is inherent and beneficial. Which isn't exactly surprising. However I don't hold out hope for evidence-based science to central to socio-economic policy, since it would force wide swaths of the population to give up thier cherished beliefs about how they wish the world works, and instead embrace how the world actually does.
Majority of players are do not express resilient behaviors, regardless of the impact on them; that is they are random, unable to create a consistent valid strategy, etc. - it is not that they are just greedy, since if that were the case, the system dynamics would be much more stable.
It was not surprising in face of what was previously known, but it was not obvious either.
A general challenge in doing social science research is that once you present a new result, someone will always claim that it was obvious because of X. What they don't see was that there was a Y that could have plausibly explained an entirely different result. Once we've shown which of X or Y is more likely, the world updates, but it was not obvious to begin with.
My collaborator Duncan Watts (one of the pioneers in the field of network science) has written a book about this phenomenon: Everything is Obvious, Once You Know the Answer.
If you're saying that the resilient players might switch to being non-resilient, sure, it's not obvious what would happen, but my assumption is that a resilient player is resilient.
Now it's obvious, because we run into nice, prosocial people and of course they fit the profile.
Anyway, I made no headway with my professors (or other students, probably.) They were happy to call the results they then had proof that tit-for-tat was tops, and always would be 'cause it was such a simple, logical strategy. So, no, I can personally assure anyone who cares to know that it wasn't obvious to the experts then. As for laypeople... well it's most accurate to say they had no opinion; it was rough trying to explain PDs back then since they violated everyone's "folk expectations" in some way, whether you were religious or a rationalist and were rarely spoken or written of outside of academia.
I'd also say I didn't know it then, either; I just suspected - since it wasn't perfectly clear that my analogy from more ambiguous real-life exchanges would hold in such clearly-defined circumstances.
Respond in kind to every action taken against you. If someone offends, attack, if they offer peace, be peaceful.
Nice people will more often offer the olive branch (sub-optimally), so anyone else playing the optimal strategy will offer it back. Over time this probably results in a lower violence rate. Even mean people (also sub-optimal) will get dragged into niceness over time since nice/mean probably changes dynamically.
( * I think it's this: https://sciencehouse.wordpress.com/2012/09/04/a-new-strategy... )