Novel response production as a function of variability and amount of practice.

Abstract This study investigated the effects of three different amounts of practice in combination with two types of variable practice conditions upon schema development. Seventy-two subjects were administered either 6, 18, or 36 trials while learning to move to either one- or three-criteria goals on the linear positioning task. Subjects then had 18 no-KR (knowledge of results) trials to produce a novel response. The statistical analysis revealed a significant three-way interaction for absolute error while no significant main or interaction effects existed for constant or variable error. Analysis of the simple main effects showed that the various amounts of practice produced similar performances for the subjects learning to move to only one criterion goal. For the subjects learning to move to three criteria goals during initial practice, the amount of practice provided was a significant factor in the accuracy and strength of the motor schema. Partial support is presented for schema theory.