Pattern Recognition in Descartes' Automata

I IN HIS TREATISE ON MAN (L'Homme, about 1632) Rene' Descartes develops an explanation of the way in which an automaton might recognize and respond to patterns in its sensory data. This explanation seems to underlie the theory of memory that appears in Descartes' major works. It is also related to his account of innate ideas and to his theory of error. As his later letters show, however, Descartes did not find this purely physical theory of pattern recognition adequate to explain human memory. He seemed to need a theory of a purely "intellectual" memory, which could not be simulated by an automaton. The story of the rise and fall of Descartes' early theory of memory is important not only for its historical interest, but also because it suggests a possible limitation upon purely physical, automaton-like theories of pattern recognition.