Aging agents

Abstract

We are adopting Brooks and Wiley's view of evolution as an irreversible process capable of producing increasingly greater complexity at higher organizational levels. We start from the assumption that the evolutionary force is intrinsic in the living system, and is in reality a continuous senescence function leading gradually and unavoidably to death. We are therefore seeking a senescence function that favors social rather than solitary agents in terms of longevity, without prespecifying in detail the agent's life span. We show that a senescence function relyling on negative (destructive) feedback links from metabolism to genetic program conforms with these specifications. We also show that senescence should affect all the regulation parameters of the agent, and that the system remains nonmanipulable and unpredictable as far as its life span is concerned. This senescence function favors the more “cognitive” agent models (the ones having additional regulation loops), and thus the emergence of organizations of a higher order that have more elaborate social relations.