Abstract

Two experiments were performed to study inductive reasoning as a set of thought processes that operat(s on the structure, as opposed to the content, of organized h.emory. The content of the reasoning consisted of inductions concerning the names cf mammals, assUmed to occupy a ruclidean space of three dimensions (size, ferocity, and humanness) in organized memory; Problems used in the experiments were analogies, series, and classifications. They were presented in unrestricted-time form by paper and pencil for-choice.items and tachistosccpically as two choice items. Time required in the tachistoscopic form was assumed to be an additive combinition nf five stages: selection of components of the problem, of strategy, of internal representations for the information, of a speed-accuracy trade off, and the monitoring of solutions. Results partially confirmed the vector model of memory space for names of mammals. Response latencies Confirffed the prediction that analogies regdired ore more processing step than did series, which in turn required one more step thr, did classifications. The significance of these findira for the exclanatior cf aeneral intelligence is.discussed. (Author/CTM)