The Use of Genetic Algorithms in Researching Non-Veridical Perception

Abstract

Synthetic approach to (cognitive) science – researching (cognitive) phenomena with computer and robot models – has been called upon by various field authorities, such as Froese, Ziemke and Harvey, to tackle the problem of opposing theories that have pestered Western philosophy for centuries, especially those of epistemic nature. One synthetic methodology can offer comparison of such theories under the mechanism of natural selection – genetic algorithm. Specifically, genetic algorithms can be deployed to research non-veridical perception, the viewpoint held by various paradigms (e.g., constructivism) that the world we experience is not a representation of the world out there. One such theory that boasts empirical proof is the interface theory of perception. However, genetic algorithms, although bearing an ecologically viable modeling platform in the form of natural selection, can be, due to yet undiscovered biological realities, largely manipulated with arbitrarily set parameters and methods to get biased results. What’s more, GA-based research on non-veridical perception does not seem to include full computational, algorithmic and implementational materials. This begs a carefully set protocol for such research.

Publication
In O. Markič et al. (Eds.), Cognitive science: proceedings of the 20th International Multiconference Information Society – IS 2017, 9th-13th October 2017, Ljubljana, Slovenia: volume B (pp. 25–28). Ljubljana: Institut Jožef Stefan
Date
Links