Turing Completeness of Neurons

• We've shown that anything that can be programmed, can be programmed with a CA net with enough neurons.
• That is, neurons are Turing complete.
• We've used this to develop neural systems, and in particular neural robots.
• So we have Cell Assembly roBots (CABots), that move around virtual environments.
• One simple thing to do is use the CAs to make finite state automata.
• We use the FSAs for planning and natural language parsing.
• However, we also use the neurons for vision.