Trading Evolution for Communication

Abstract

In this chapter we still deal with membrane systems which have multisets of symbol—objects in their regions, but with the evolution rules of a type much different from the rewriting-like rules considered in the previous chapter, and closer to ways of transferring chemicals through membranes as encountered in biology. Specifically, we will compute by communication only, by changing the places of objects with respect to the membranes of a system, and not by changing the objects themselves. We will find that also in this framework we can get computational universality, and this is a rather interesting result, from various points of view. First, as we will see, the rules we use correspond directly to well-known biochemical processes. Second, because we do not change the objects, we do not create and we do not destroy them, and this means that the systems which follow observe the conservation law (which is not necessarily the case with the systems considered in the previous chapter, where rules of the form a → aa were used without taking care of the way of producing two copies of an object from only one existing copy). Third, the universality results show that in our framework communication is rather powerful. Communication makes the difference between a collection of separate computing agents and a system of cooperating agents; it is related to the so-called “system effect” (to synergy, emergent behaviour), so it is somewhat a common sense statement that communication is powerful.