Perspectives for Applied Semiotics in Artificial Life Research (original) (raw)

The contributions of Peircean pragmatic theory of signs to the design and construction of artificial cognition systems have not been systematically explored. In fact, most approaches in the literature of intelligent systems and artificial life adopt a naïve definition of semiotic processes, which usually plays a secondary role in the studies. Our research, on the contrary, strives for a strong theoretical foundation for semiosis, as well as its realization within digital computers. In this lecture, a biologically inspired semiotic model is proposed in synthetic biology. At the first part of this lecture we investigate theoretical constraints about the feasibility of simulated semiosis. These constraints, which are basic requirements for the simulation of semiosis, refer to the synthesis of irreducible triadic relations (Sign – Object – Interpretant). We examine the organization of the triad S-O-I, that is, the relative position of its elements and how they relate to each other by determinative relations, and we suggest a meta-algorithm. At the second part we begin with a description of a general approach for conducting experiments with artificial creatures within a synthetic ethological context. Next, we describe how this approach was used to build a computational experiment regarding the emergence of self-organized symbols. Our experiment simulated a community of artificial creatures undergoing complex intra and inter-specific interactions in which meaning evolved over time, from a tabula rasa repertoire of random alarm-calls to a specific set of optimal referential alarm-calls. To design different kinds of creatures as well as innanimate elements of the environment, we applied theoretical constraints from the Peircean philosophy of sign and empirical constraints from neuroethology. Behaviors such as navigation, search, predation, evasion and cooperation were modeled as communication processes evolving within and across artificial brains of different kinds of creatures. Our results suggest that the constraints chosen were both necessary and sufficient to produce symbolic communication.