Self-modifying systems: a model for the constructive origin of information.

The mechanisms of information processing in Turning machines and biological systems are examined from the point of view of physical sets of variables. Computation is characterized as a process, the realization of which involves a bounded set of interactions and a pre-definable set of variables in real systems. Using ideas from process philosophy, the ability of natural systems to transcend these computational modes is discussed. A class of systems, called self-modifying systems, that utilize persistent shifts in their defining interactions and variable composition is introduced. Various other ideas that lead to similarly non-computational or semi-computational scenarios (as in the case of distributed code systems) are referenced. As applications, computer models of emergent phenomena using randomly growing interaction sets as well as theoretical issues that range from the meaning of simulation to the problem of information gain in self-modifying systems are discussed.