Generalization of serial position in rote serial learning.

Publisher Website
Google Scholar
Abstract
To investigate the notion, proposed by Underwood, that there is differential generalization among serial positions, 150 S''s were asked to learn a 10-item serial nonsense syllable list under 5 different conditions. Level of list mastery was varied in 3 ways; intralist similarity in 2. Immediately after learning subject was asked to designate the ordinal number of the serial position occupied by each syllable in the list he had just learned. Inaccuracy in naming these positions provided the operational definition of generalization of serial position. Empirically, the following was established: (1) Accuracy in position naming varied as a function of serial position. The function was bow shaped, with a particular position''s relative difficulty remaining fairly constant even under widely varied conditions of learning. (2) Knowledge of serial position per se reached its maximum level of perfection somewhere between 5 out of 10 correct and 1 perfect plus 5 trials in terms of list mastery criterion. (3) Formal intralist similarity, though affecting position naming, cannot be held entirely accountable for the results of the experiment. These findings support the generalization hypothesis in terms of the definition set forth. However, certain other findings made this definition somewhat tenuous, so it was necessary to conclude that the experiment did not adequately evaluate the hypothesis of generalization of serial position.
Keywords

This publication has 0 references indexed in Scilit: