Computers and Thought Lecture: The Ubiquity of Discovery

Abstract

\s scientists interested in studying the phenomenon of "intelligence", we first choose a view of Man, develop a tl!eory of how intelligenf behavior is man;~ g cd, and construct some mndcls which Gill test out and reline that theory. T he view we choose is that Man is a symbolic infonnrttion processor . The theory is th<~t sophisticated cognitive t;~sks can. be cast i\S searc h es Or explorations, illld that each . human possesses (and efficiently accesses) a large body of informal rules of thumb (l?euristics) which con.strain hi s sea rc h . The so ur ce of what we colloquially call "intelli gence" is seen to be very efficient searching of · an a fniori immense sp<~ce. Some computational models which incorpor<1te. this theory arc · described. Among them is AM, a cornputer pro g ram which develops new mathcm;~tical concepts and conjectures involving thr.m; AM is guided in this exploration by a collection of 250 more or. less general heuristic rul es. The operational natur e of .such models . allows P.xperimcnts fo be performed upon them, experime nts which help us test and develop hypotl1eses about intelligence . One interestin g result ha.s been the ubiquity of this kind of heuristic g uid ;~ n ce: intelligence pcrme;~tr.s everyday problem solving <1nd inv ention , as well as the kind of problem solving and invention that scientists and artists perform.