Unfortunately, game theory does not provide the magic formulas that allow you to win at poker or chess, not even at the stock market, even though its main applications are in economics. And although John von Neumann, father of the idea, was inspired by the strategies that guide the performance of good poker players, Reinhard Selten, his follower, assures that he never plays cards and that he does very badly at chess.
For the layman, delving into this discipline is difficult, and Selten does not help much to reveal what it consists of: “It is impossible to explain it in five minutes, but the definition I give when asked is that it is a mathematical analysis that models the competing conflicts “.
Despite everything, the conference he recently delivered in Madrid, invited by the Debate Club of the Complutense University, allowed the audience to get a broad idea of his work.
– In addition to mathematical formulas, does game theory use elements of psychology?
-Some branches of game theory take into account contributions from psychology, but the main body does not, because it only studies the interests of people and tries to explain what each one does based on that interest and the circumstances . If you know that someone wants to travel to Barcelona and that they do not have a car or like to fly, you do not need any psychological element to predict with a certain degree of security that they will go by train.
– The great strategists of history, such as Alexander, Caesar or Napoleon, weren’t using a similar analysis, even if it was by intuition?
-In some respects, it is possible. People do many things by intuition that cannot be explained in a systematic way. Game theory does not model intuitive behaviors, although the great strategists were able to make reasonings similar to those of this discipline. We simply try to analyze the interactions between purposeful people in situations of conflict and cooperation: what is more reasonable for them to do and how to prevent the steps they are going to take.
– I suppose there is a big difference between the situations in which all players can win or lose and those where the victory of one means the defeat of the other …
-Sure. In the latter case we call it a zero-sum game, because what one wins is lost by another or others, but it is not the most usual case.
–Although there are many games of this type, such as poker, chess …
-Yes, but they are not the ideal field of application of the theory, because competitive efforts are divided here. In real life, these cases are usually much rarer.
–Wouldn’t war also be a zero-sum game?
-Many wars end with the unconditional surrender of a contender, but in others, as has happened recently in Yugoslavia, it is necessary to reach an agreement in which everyone gives up something and everyone gains a lot. The strategy of the Croatian offensive last summer made it possible to vary the positions of each side and facilitated the development of the peace negotiations.
–Was the theory then used?
-I don’t know, but, of course, the whole process could be analyzed with modeling coming from theory.
–Where does the name of game theory come from?
-From the first article that was published on the subject, that of John von Neumann. At first, the theory studied these zero-sum games, in which one can only win what another loses. So the name is only historical, as it is with so many others in science. Some people have proposed changing the name to something like interactive decision theory.
–Does game theory offer solutions to a problem or is it just an analytical tool?
-It is a tool for analysis and, in some cases, it can give certain recommendations on what to do in a specific situation; but this is rare, because real situations are too complex to model completely. It is not easy to give practical advice.
–Is the field of greatest application found in the economy?
-It has served to analyze very specific cases, such as business strategies in an oligopoly situation, the breakdown of a monopoly or the choice of investments.
–Do you think that companies and governments really use this tool before making their decisions?
-It is difficult to know, because those who do keep it a secret. It is not something that companies talk about, because it would be giving clues to competitors about what one intends to do. Also, all decisions must be based on different tools. If the government, for example, decides to devalue the currency, it will have to consider the reactions this will elicit in the markets and the actions that companies and other governments will take. Game theory will help you perform this analysis, but it will not give you the exact percentage of devaluation.
–And in science, what applications does it have?
-In the United States, there is a researcher who is applying game theory to the study of aggressive behavior in certain species of spiders: how this behavior is shown depending on each situation. (See the case of the birds of paradise in Larry Gonick’s Humor, pages 145 and 146 of this same issue.)
–It must also be important in ecology, since it studies equilibrium and disequilibrium relationships, right?
-It can be important, up to a certain point, to understand how to reach the ecological balance and the struggle of the species that form it. And this includes the study of biological evolution. It is studied how environmental changes condition the response of a species and survival strategies. For example, the pollination of flowers by bees is a strategy that works so successfully that it has conditioned the evolution of flowering species and bees.
–And is it applied in medicine?
-It is used in the relationships between doctor and patient, although this is more social than scientific. The doctor must establish a strategy to convince the patient of the need to follow his treatment accurately. The laboratories take this into account and try to facilitate the ingestion of the pills by modifying the doses so that the treatment is as simple as possible for the patient. It is also useful for questions of health economics or the health system, where many factors come into play.
–In some way, all the relationships in our daily lives – at work, at school, with the neighborhood, in marriage, etc. -they are a game in which it is possible to apply the theory …
-You could think so. Knowledge of game theory allows one to deepen understanding of relationships and prevent unwanted situations. But it cannot be said that everything is the subject of game theory. The dilemmas and situations of everyday life are usually not so complex as to require mathematical modeling. It is true that it could be said that everything is a game, but in the same way that it can be said that everything is physics, that everything is chemistry or that everything is economics.
Ignacio F. Bayo
This interview was published in May 1996, in number 180 of VERY Interesting
