Here is another short tantalizing tidbit of creative genius in the world.  Jun Tani, Senior Researcher with Sony's Computer Science Lab (CSL) in Japan, is exploring neural-net robotic systems.  On the face of it, that may seem rather mechanical and clinical, until you realize that his premise is the organization of allowing information processors to be accessible and adaptive to the full spectrum of their environmental encounters.  The systems are allowed to have free-floating programming and come to thriving balances with the world they continually discover (not just function in).

Jun has informed me that full texts of his work are available through his website at:

                                           http://www.csl.sony.co.jp/person/tani.html

(1/10/98)

June, 1997                                Claremont College

Presentation Abstract     

Jun Tani
Sony CSL

THE BOUNDARY IS SELF-ORGANIZED BETWEEN IMPLICIT/EXPLICIT COGNITION:
A ROBOT LEARNING EXPERIMENT

We describe dynamical phenomena which resulted in robot learning experiments and interpret their meanings in consciousness problems.  We observed that global attractor of fractal is self-organized in neural dynamics during successful learning. Rationality is kept when the internal state transits within this attractor, which corresponds to explicit cognition. Rationality can be lost when the state is perturved to the outside, but it is always recoverd by means of global attractor dynamics. This corresponds to implicit cognition.