Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!aramis.rutgers.edu!athos.rutgers.edu!nanotech From: peb@tma1.sun.com (Paul Baclaski) Newsgroups: sci.nanotech Subject: Re: Consiousness, the body, and the soul (long and mystical) Message-ID: Date: 13 Jul 89 03:35:06 GMT Sender: nanotech@athos.rutgers.edu Lines: 29 Approved: nanotech@aramis.rutgers.edu In article , yale!root@seismo.css.gov (Root Of All Evil) writes: > > In article macleod@drivax.UUCP (MacLeod) writes: > I have thought of a model of thought processes where the thought > processes are represented by a curve in some high-dimensional space (I > haven't classified the dimensions), with one dimension being time. A > person can be specified by giving one point (a memory state at one > instant in time) and the derivative of the curve... Interesting model, but I suggest you consider that not all functions have a derivative. You might want to update your model by studying non-linear dynamics (chaos). Paul E. Baclaski Sun Microsystems peb@sun.com [And even if it does, the derivative is likely to be a function as complex as the original function. The derivative *at the point* won't do you any good. Consider that there are an infinity of functions that go through the point (3,7) and have the derivative 1.5 there. --However-- there is a model useful for other things that is essentially this, called BSB (Brain State in a Box) developed by J. A. Anderson. See Laberge & Samuels (eds) "Basic Processes in Reading Perception and Comprehension", pp 27-90 (Erlbaum: Hillsdale NJ) and Rummelhart and McClelland, (eds) "Parallel Distributed Processing", v1pp66-68 (MIT Press). --JoSH]