Slime mold intelligence

Computer science searches in biological systems for ideas which might by modeled in silico, e.g. artificial intelligence (AI),  genetic algorithms (GA), or neural nets. Hitherto the progress was somewhat disappointing. Like the rule based classical AI which fades away. Or GA which approach solutions only asymptotically.  The reason for such inadequacies, is an adherence to anthropomorphism. Robots are supposed to mimic humans, even think and feel like humans. To me this avenue is barren, and  time has come to search for simpler alternatives which are more promising.

Let’s start with an important insight by the great philosopher Hans Jonas. “The emergence of the human mind does not mark a great divide within nature but elaborates  what is prefigured throughout the life-world.” (The Phenomenon of Life ISBN 0-8101-1749-5).This is the most important conclusion of evolution theory. Since evolution is a continuous process, the mind has also  evolved in a continuous fashion. Not only the mind but anything which makes us human, e.g., memory, intelligence and creativity.  

Computer scientists ought to turn their attention to  simple organisms like an ameba  called Physarum polycephalum . When this ameba  is subjected to a series of shocks at regular intervals, it learns the pattern and changes its behavior in anticipation of the next one to come, according a team of Japanese scientists (Nature 451, 385 (2008)). Toshiyuki Nakagaki of Hokkaido University in Sapporo and his colleagues say that their findings “hint at the cellular origins of primitive intelligence”.( Saigusa, T., Tero, A., Nakagaki, T. & Kuramoto, Y. Phys. Rev. Lett. 100, 018101 (2008)).

Since lacking a nervous system amebas  memory is embodied. And so is their intelligence, manifested by their creative response to injury. These experiments suggest that amebas may be trained in a Pavlovian fashion,  although this was not the research objective. Why not create simple objects (robots) with ameba intelligence? Which may be achieved with CA.   

v. CA  memory
v. CA  embodiment