Effect of number of choices per unit of a verbal maze on learning and serial position errors.

Publisher Website
Google Scholar
Abstract
Five groups of Ss learned a 16-unit verbal maze for which the number of alternate choices per unit were 2,3,4,6, or 8. Rate of acquisition measured in terms of mean total time and mean total errors to the criterion of one perfect repetition increased with increase in the number of alternate choices per point. Analysis of variance indicates a significant effect of the independent variable on both measures of acquisition. Tests of deviation from linear regression reveal no deviation. Therefore, straight line regression equations were fitted to the data by the method of least squares. These functions are y=34.1x [long dash] 12.8 for errors and y=115.4x+157.4 for time in seconds. Theoretical discussion of these functions suggests that in the determination of maze difficulty the effect of number of alternate choices per choice point occurs early in acquisition; the effect of number of like responses in the correct pathway occurs later in acquisition; and the number of like responses is primarily responsible for the disproportionate decrease in the increments of maze difficulty. Serial position error curves were constructed from the data for each group, and extent of bowing of these curves was found to increase with increase in the number of alternate choices per choice point. An analysis of variance shows this relation is significant statistically. Other characteristics of the serial position error curves are comparable to those obtained in studies of serial verbal learning and maze learning.
Keywords

This publication has 0 references indexed in Scilit: