The spatial relations among touch sensors on the skin is mirrored in the brain. Because of this, "in general, the greater the extent to which two stimuli arouse the same sensors, the more nearly alike will be the partial metnal states those stimuli produce -- and the more similar those stimuli will `seem' simply because they'll tend to lead to similar mental consequences."
"Other things being equal, the apparent similarity of two stimuli will depend on the extent to which they lead to similar activities in other agencies." (NB: This implies a very strong relationship between brain states and mental states.)
Provided that there is spatial regularity in sensors, agents can learn topographic maps. The spatial relationships begin imposing constraints on the kind of inferences (or representations) that can be drawn by higher-order agents.
(NB: First, doesn't this remind you a lot about Braitenberg's vehicles? Second, it makes me think quite a bit about the natural computation approach to vision, where certain natural constraints concerning the physical world (e.g., 2D projected topography) are mirrored in the early stages of transducers.)
Pearl Street | Society of Mind Home Page | Dawson Home Page |