Hypothesis behavior as a function of amount of pretraining.

Publisher Website
Google Scholar
Abstract
Performed 2 experiments to determine the effects of conceptual pretraining on the hypothesis behavior exhibited by human Ss on simultaneous discrimination problems that included blank trials. For Exp I, 80 Ss were given 0, 1, 2, 3, or 4 simultaneous discrimination problems prior to 4 blank trials problems. Exp II was defined by the factorial combination of 0, 2, 4, or 6 pretraining problems and 2 sets of instructions. 128 undergraduates were given routine instructions or directions to consider a single hypothesis at a time. With training, Ss were more likely to maintain their previous hypothesis after positive outcomes and to alter their hypothesis after negative outcomes. It is concluded that the uniprocess hypothesis models (e.g., M. Levine's 1966 model) do not accurately describe the behavior of untrained Ss, and that a multiprocess model of conceptual learning is needed. (PsycINFO Database Record (c) 2006 APA, all rights reserved)

This publication has 0 references indexed in Scilit: