Selfish driving causes everyone to pay the Price of Anarchy
While the result of this study is not surprising, the fact that this study was done and the methods it used are really interesting.
An illuminating observation that the study brings is the "Nash equilibrium" (a point at which no benefit will be realized if one's strategy is changed unilaterally) is reached when more and more people move away from the global optimum. What this means is that if a single person decides to start playing by the rules on the roads, he will still be late if the people around him don't follow the rules. The whole setup needs to change to benefit everyone. Sample this quote from the paper's abstract:
In the book "Games Indians play" by V Raghunathan, the author touches upon this point. But he states that this condition is unique to India. However, this study by scientists from KAIST (Korea Advanced Institute of Science and Technology) while focusing on traffic shows that this is essentially a "human" problem.
The system cannot be optimized for maximum possible individual benefit without substantial costs. Optimization for maximum system benefit is what is needed; which will translate to significant individual benefit as well. There are plenty of team games that illustrate the same point quite well.
I was reminded of a set of brilliant essays (Collection titled "The Quest After Perfection") I am reading now by none other than the philosopher-sage Professor M Hiriyanna. In his essay on the "Ethics of the Upanishads", he defines evil to be two fold (I am paraphrasing here).
1. The urge for instant self-gratification as opposed to deferred pleasure.
2. The prioritization of one's own pleasure over that of society and the environment.
To be not-evil, therefore, one should have deferred gratification of pleasure and keep societal needs before the individual's, which is essentially the same point the authors of the paper make.
Dharma is a central Indian concept that involves putting society before one's self. The idea here is that it will make life easy not just for yourself but the entire society, while benefiting each individual finally.
In this age, individual liberties are held to be more important than societal good. The assumption behind this seems to be that a bunch of individually happy people will make society happy as a whole. Unfortunately, studies such as the one referenced show that individual maxima do not translate to a system maximum.
The questions that arise in my mind are:
1. To what extent can individuals forego "personally optimal strategies" to benefit society as a whole?
2. Which one is preferred? A non-optimal society where people have a little more "freedom" to pursue personally optimal strategies? Or an optimal society which, by design, will be less "free" to pursue personally optimal strategies? Is there a meeting place for the two?
The paper itself can be found here.
While the result of this study is not surprising, the fact that this study was done and the methods it used are really interesting.
An illuminating observation that the study brings is the "Nash equilibrium" (a point at which no benefit will be realized if one's strategy is changed unilaterally) is reached when more and more people move away from the global optimum. What this means is that if a single person decides to start playing by the rules on the roads, he will still be late if the people around him don't follow the rules. The whole setup needs to change to benefit everyone. Sample this quote from the paper's abstract:
Uncoordinated individuals in human society pursuing their personally optimal strategies do not always achieve the social optimum, the most beneficial state to the society as a whole.In India, and Bangalore especially, driving is little short of anarchy. In my earlier grouses on the matter, I have consistently held the opinion that people need to pay heed to the law to benefit not only themselves but the entire society and indirectly themselves as well.
In the book "Games Indians play" by V Raghunathan, the author touches upon this point. But he states that this condition is unique to India. However, this study by scientists from KAIST (Korea Advanced Institute of Science and Technology) while focusing on traffic shows that this is essentially a "human" problem.
The system cannot be optimized for maximum possible individual benefit without substantial costs. Optimization for maximum system benefit is what is needed; which will translate to significant individual benefit as well. There are plenty of team games that illustrate the same point quite well.
I was reminded of a set of brilliant essays (Collection titled "The Quest After Perfection") I am reading now by none other than the philosopher-sage Professor M Hiriyanna. In his essay on the "Ethics of the Upanishads", he defines evil to be two fold (I am paraphrasing here).
1. The urge for instant self-gratification as opposed to deferred pleasure.
2. The prioritization of one's own pleasure over that of society and the environment.
To be not-evil, therefore, one should have deferred gratification of pleasure and keep societal needs before the individual's, which is essentially the same point the authors of the paper make.
Dharma is a central Indian concept that involves putting society before one's self. The idea here is that it will make life easy not just for yourself but the entire society, while benefiting each individual finally.
In this age, individual liberties are held to be more important than societal good. The assumption behind this seems to be that a bunch of individually happy people will make society happy as a whole. Unfortunately, studies such as the one referenced show that individual maxima do not translate to a system maximum.
The questions that arise in my mind are:
1. To what extent can individuals forego "personally optimal strategies" to benefit society as a whole?
2. Which one is preferred? A non-optimal society where people have a little more "freedom" to pursue personally optimal strategies? Or an optimal society which, by design, will be less "free" to pursue personally optimal strategies? Is there a meeting place for the two?
The paper itself can be found here.
