Information processing in multigame environments modeling the evolution of sex via punctuated equilibria

Uncertain environments are properly described by probability distributions which, as usual, can be collapsed or conditioned into distributions with reduced uncertainty through the processing of environmental information. Organisms which force this collapse gain evolutionary advantage by being able to employ strategies in a known environment rather than in a merely probable one. The accrued benefit gained from processing information can be precisely quantified by comparing benefits returned using distributions prior to, and after collapse, and these often large and immediate benefits can amply justify the evolutionary cost of information processing systems. More importantly, the evolution of information processing systems must necessarily occur in a predictable evolutionary sequence from less complex to more complex. Practical applications include modeling the evolution of sex modeled here as a sequence from asexual reproduction, to single gene exchange, to gene packet exchange, to same species packet exchange, and finally to sexual reproduction, sexual selection, "Red Queen" contests and so on. Modeling this sequence requires extensions to game theory originally designed to model a single game, to allow the simultaneous operation of many games. This extension is called a multigame environment. The dynamical evolution of the development sequence shows punctuated equilibria. (C) 2000 Academic Press.