Dynamics and Evolutionary Games:

The  theory  of  games  is  based on  the  assumptions of rationality and equilibrium has proved  to be very usehl, but  it would  be a mistake to rely on it  totally. When games are played  by  novices who do not have  the necessary experience  to  perform  the calculations to  their  best  strategies, explicitly  or implicitly,  their choices and  therefore  the  outcome  of  the  game, can differ significantly from  the  predictions  of  analysis  based  on  the concept  of equilibrium.

However, we  should not abandon all notion of good choice, we  should allow for  a  dynamic process  in  which strategies that prove  to  be  better  in  the previous plays of the game are more likely to be chosen in the later plays. The evolutionary approach to games does  just this. It is derived  from  the  idea of evolution  in  Biology. Any  individual animal's  genes strongly influence its behaviour. Some behaviour succeeds better  in  the prevailing environment,  in the sense that animals exhibiting those behaviors are more  likely to reproduce successfully  and  pass their  genes  to their progeny. An  evolutionary  stable state, relative to a given environment, is the ultimate outcome of this process over several generations.

The analogy  in  games would  be  to  suppose that strategies are  not  chosen  by conscious, rational maximisers,  but  instead that each player  comes  to  the game with a particular strategy "hardwired"  or "programmed"  in. The players then  confront other players who  may  be  programmed  to  play  the same  or different strategies. The  strategies that are fair better get higher payoffs and multiply faster while the other decline.  In  the  context  of  strategic  games  in business  and  society, the mechanism is  much  more likely  to  be  social  or cultural -  observation and initiation, teaching and learning, greater availability of capital for the more successful ventures and so on. 

The  object  of  study  is  the  dynamics  of  this  process. Interestingly  in  many games, the evolutionary stable limit  is the same as the equilibrium that would result  if  the  players were conscious, rational calculators. Therefore,  the evolutionary approach gives  us  the  backdoor  justification  for  equilibrium analysis.