Computational
Intelligence and Semiotics –
on the synthesis of semiosis in artificial systems and creatures
João Queiroz[*] & Ricardo Gudwin
© This paper is not for reproduction without
permission of the author(s).
Synthetic methodologies have been used to model and simulate semiotic processes from many different perspectives. Computational Neuroethology, Artificial Life, Animats, Synthetic Ethology, and Computational Semiotics are some of the interdisciplinary areas of research involved in the synthetic design of semiotic systems and creatures. These areas have been designing artificial environments that work as experimental labs, where it is possible to test the predictions derived from theoretical models (Bedau 1998, Parisi 2001), and provide us with opportunities to specify theories with computational formalisms (Parisi & Cangelosi 2002).
Depending on the framework, the synthetic based strategies permit the testing of various factors affecting semiotic processes, such as the differences between innate and learned sign systems, the adaptive role of compositional languages, the adaptive advantage of symbolic processes, the hypothetical neural substrate of these processes, the mutual influences between different semiotic competencies and low level cognitive tasks (e.g., attention, perceptual categorization, motor skill), and the hierarchical presupposition of fundamental kinds of semiotic competencies (e.g., indexical → symbolic).
This S.E.E.D. Special Issue puts together researchers from many fields to address some important questions: What are semiotic and symbolic processes? What is a semiotic machine? What kind of theoretical and empirical constraints must we consider to simulate semiosis? Can an evolutionary computational approach to semiosis reveal the mechanisms involved in symbolic competence emergence and performance? Is the semiotic behaviour an emergent property resulting from the dynamical interaction between an embodied creature and the environment? How can higher level semiotic processes can emerge from lower level ones?
This Issue is the outcome of the II Workshop on Computational Intelligence and Semiotics, happened in August 2002, at Itaú Cultural, São Paulo, Brasil. The meeting was supported by the State University of Campinas (UNICAMP) and the Pontifical Catholic University of São Paulo (PUC-SP), and was sponsored by CAPES, FAPESP, and Itaú Cultural. The workshop organization format privileged open discussion and debate from theoretical issues to applied intelligent system software implementations. Researchers from computer science (Tom Ziemke), engineering (Leandro de Castro, Ricardo Gudwin, Ângelo Loula), cognitive science (Pim Haselager, Maria Eunice Gonzalez), neuroscience (Sidarta Ribeiro, Ivan Araújo), philosophy (Andre De Tienne, Joseph Ransdell), computational linguistics (Alexander Mehler), linguistics and semiotics (Winfried Noth, Lucia Santaella, João Queiroz) were invited.
DeTienne examined the Peircean concept of learning as a time embedded semiotic process regulated by the principles of the logical phenomenological category of thirdness. He established five general constraints according to which Peirce understands learning as a pre-psychological phenomena. Lúcia Santaella focused on Peirce’s concept of symbolic process, a controversial topic in Cognitive Science, which assumes that a symbol is a very structured entity, frequently in a declarative system of signs, a property to which Peirce did not restrict this concept.
Joseph Ransdell developed Skagestad’s distinction between the complementary fields of Artificial Intelligence or “AI” and Intelligence Augmentation or “IA”, with emphasis on the dialogical aspects of embodied sign processes. He analysed a case study of intelligence augmentation where computational techniques were used for performing the critical control of scientific communications and publications, focusing on the computational assistance of communicational practices involved in research activity. Maria Eunice Gonzalez and Pim Haselager argued on the contribution that the concept of abductive reasoning as a self-organized process can make to cognitive science investigations on creativity. They also discussed the limits for the implementation of abduction in artificial machines. Winfried Nöth examined, in the light of Peirce’s concept of genuine and quasi semiosis, the notion of the “semiotic machine”. According to his approach, a pragmatic dimension of situated and self-regulated creative processes can possibly indicate a threshold between a quasi and a genuine semiotic machine.
Alexander Mehler discussed methodological issues regarding foundational aspects of computational semiotics. He distinguished among different approaches within computational semiotics, which he called "computer semiotics", "artificial semiosis" and "the cosine approach" (computational semiotics in the narrow sense), and the different modeling roles performed by each of them (in terms of simulation, realization and emulation of semiosis). Pim Haselager questioned the assumption of Umwelt as a life-dependent phenomena in the context of situated-embodied cognitive robotics. Tom Ziemke discussed the use of autonomous agents as models of sign processes and embodied enactive cognition, addressing the question whether, or to what extent, such agents are themselves autonomous and capable of semiosis, in the sense that living organisms are. Leandro de Castro suggested that new developments in artificial immune systems may have deep consequences for the research on Computational Intelligence. Ângelo Loula and colleagues proposed, based on Peircean semiotics and informed by neuroethological constraints, a methodology to simulate the emergence of symbolic predator-warning communication among artificial creatures in a virtual world of predatory events. They are particularly interested is how higher level semiotic processes emerge from lower level ones.
ACKNOWLEDGEMENTS
The authors wish to thank Julia Itani, Bia, Phillip, Thais Paparela, Paulo Schiavon, Tatiana, Luciane Rodrigues, Ângelo Loula, and Renata Pedrosa for technical assistance. Funding for the II Workshop on Computational Intelligence and Semiotics was received from FAPESP (#02/06640-7), as well as from CAPES/PAEP (#0235/02-4). J.Q. is funded by a grant from FAPESP (#02/09763-2). R.G. is funded by a grant from CNPq (#300123/99-0).
References
Bedau, M. A. 1998. Philosophical content and method of artificial life, in T.W.Bynum & J.H. Moor, eds., The Digital Phoenix: How Computers are Changing Philosophy, pp.135-152. Blackwell Publishers.
Parisi, D. 2001. Simulazioni – la realtà rifatta nel computer. Il Mulino.
Parisi, D. & Cangelosi, A. 2002. A unified simulation scenario for language development, evolution and historical change, in A. Cangelosi and D. Parisi, eds., Simulating the Evolutions of Language, pp.255-329. Springer Verlag.