Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

nature.com

Post reply on HN
Print view View original

31–40 of 57 posts

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#31

Post by dang »

coderdude wrote: post #13
andybak wrote: post #12

Earlier quoted context omitted.

Maybe stick around here but don't click on things that don't interest you? I think the Prisoner's Dilemma is one of the most interesting abstract ideas I've come across and it feeds into my thoughts about morality and human society. My interest in it is because I think it matters I fully admit that I probably "don't know what I'm talking about" but it's not for want of trying. My interest is sincere. Is your cynicism…

I don't think you're the target of my cynicism. I think you're a dick.

We've banned this account for violating the HN guidelines.

If you don't want it to be banned, you're welcome to email hn@ycombinator.com and make this right. We don't like to ban longstanding participants, but you know HN well enough to know that you can't comment like this here.

https://news.ycombinator.com/newsguidelines.html

https://news.ycombinator.com/newswelcome.html

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#32

Post by espeed »

mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

What if you modify the game a bit...?

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.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#33

Post by mizzao »

espeed wrote: post #32
mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

What if you modify the game a bit...? 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: Coope…

Really cool suggestions - they actually encompass a couple of ideas that we've had.

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.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#34

Post by mizzao »

shawnee_ wrote: post #29
mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

I wonder if a better one sentence conclusion might be something like "Predictably nice players consistently distribute their gains to the group". After all, the interesting variable here is repeatability; that is, as certain players learn which players are consistently "nice" through repeated trials, they get iteratively better at gaming them to forfeit their (potential) gains. In other words, over time, the "compoun…

It is true in our experiment that there are players who are consistently nice, and they experience lower daily payoffs than the other players. So, in one sense, "good guys finish last".

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.)

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#35

Post by mizzao »

nabla9 wrote: post #30
mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

Has anyone tried to test test how different kinds of prejudges affect these games? 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.

Priming effects definitely have a huge effect on initial PD play. In one experiment researchers phrased it as "the cooperation game" vs "the wall street game" and saw much lower cooperation.

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.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#36

Post by mizzao »

vmarsy wrote: post #26
mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

Hi, interesting findings! 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

Yes, in our paper, we identified the "nice" group versus the other group. The nice group is significantly worse off, but (see https://news.ycombinator.com/item?id=13509853) still better off than they would be in the world if they weren't being nice.

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.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#37

Post by mizzao »

ianai wrote: post #27
mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

But what cost do they incur? Is it made up for in some way to them?

See my comment at https://news.ycombinator.com/item?id=13509853.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#38

Post by mizzao »

saycheese wrote: post #25
mizzao wrote: post #24

Earlier quoted context omitted.

When designing this experiment, we were pretty sure that cooperation would break down altogether based on what people had previously tried. That it didn't and resembles more what we see in the real world (things generally work and people don't stab each other in the back) was not obvious. A general challenge in doing social science research is that once you present a new result, someone will always claim that it was…

Realize that saying something is "obvious" is often not the case, though by definition resilience adds stability and in a dynamic system any addition of any truly resilient element is amplifier of stability, especially in a finite system. 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.

It was not obvious that resilient cooperative players even existed before we conducted this experiment.

Now it's obvious, because we run into nice, prosocial people and of course they fit the profile.

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#39

Post by espeed »

mizzao wrote: post #33
espeed wrote: post #32

Earlier quoted context omitted.

What if you modify the game a bit...? 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: Coope…

Really cool suggestions - they actually encompass a couple of ideas that we've had. 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…

Are you familiar with Jonathan Haidt's work on the psychology of morality and the formation of superorganisms?

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

Re: Resilient cooperators stabilize long-run cooperation in Prisoner’s Dilemma

#40

Post by diziet »

mizzao wrote: post #21

Lead author of the paper here; AMA. The main novelty about this work is that we had 100 people play repeated PD for a month instead of The findings can be summarized in a sentence as "a minority of nice people can make everyone better off."

Were conditional cooperators playing a tit-for-tat strategy or something different? I see you citing Axelrod and Rapoport but after reading through most of the supplementary material exit questions, I most often saw descriptions similar to “I chose to cooperate unless the other person chose the defect at which point I chose to defect too for the rest of the game.”

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.

Post reply on HN