Abstract

This paper reports the application of new methods for detecting computation in nonlinear processes to a simple evolutionary model that allows us to directly address these questions. The main result is the evolutionary discovery of methods for emergent global computation in a spatially distributed system consisting of locally interacting processors. We use the general term "emergent computation" to describe the appearance of global information-processing in such systems (cf. (6,7)). Our goal is to understand the mechanisms by which evolution can discover methods of emergent computation. We are studying this question in a theoretical framework that, while simplified, still captures the essence of the phenomena of interest. This framework requires (i) an idealized class of decentralized system in which global information-processing can arise from the actions of simple, locally-connected units; (ii) a computational task that necessitates global information processing; and (iii) an idealized computational model of evolution. One of the simplest systems in which emergent computation can be studied is a onedimensional binary-state cellular automaton (CA) (8) --- a one-dimensional spatial lattice of

Citations