Date of Award


Degree Type


Degree Name

Bachelor of Arts Honours


Faculty of Community Services, Education and Social Sciences

First Advisor

Dr Craig Speelman


When people practice a task, their performance in terms of speed and accuracy normally improves in a smooth manner that follows a power function. The consistency with which the performance of a wide range of skills conforms to this power function relationship is known as the Power Law of Learning, and has been an important assumption of many of the dominant theories of skill acquisition and transfer. As such, the form of the power function that is derived from the training process has been used to predict performance when task conditions remain constant. However, Speelman and Kirsner (under review) have found that performance can be disrupted by any change in a task, especially with regard to task complexity. The present study sought to more closely examine the nature of the disruption by focusing on whether it related to a change in the conceptual requirements of the task as a function of complexity. Sixty participants were used to examine three aspects of conceptual change on an arithmetic test task consisting of problems drawn from the Six-Times table, all of which could be solved by memory retrieval. The three respective conditions dealt with the influence on the test task of: other memory retrieval problems; algorithmic processing; and a combination of memory retrieval and algorithmic processing. All participants received a common set of 72 test task problems in the training phase of the experiment, which were combined with 72 distractor task problems and presented to participants in exactly the same manner in the transfer phase. Mean reaction times and accuracy were measured for the test and distractor tasks in the training and transfer phases. An analysis of the results supported Speelman and Kirner's findings of a disruption in performance at the introduction of the distractor task problems. Such a disruption was found in the conditions examining algorithmic processing and a combination of memory retrieval and algorithmic processing. In contrast to Speelman and Kirsner's findings, where participants quickly recovered from the disruption, a prolonged disruption was found in all conditions that continued throughout the transfer phase. These findings, together with those of Speelman and Kirsner, challenge some of the assumptions of current theories of skill acquisition and transfer, and provide further avenues of research into the factors affecting the use of skills in a new environment.