ABSTRACT
In structural learning, it appears to be relatively much easier to identify and/or manipulate relevant knowledge. This chapter discusses some basic questions on structural learning. Decision making capabilities correspond to decision rules and refer to the ability to determine whether or not any one of a class of entities satisfies some condition. In most structural learning, goal conditions that are intrinsic to the task itself appear to be the rule. The experimental literature on learning, however, deals primarily with experimenter-determined conditions, conditions which are usually defined in terms of extrinsic reinforcement. A major purpose of the ensuing discussion is to show how the general point of view represented by the simple performance mechanism may lead to: useful predictions about complex human behaviour; and other assumptions which may play an even more crucial role in cognitive theorizing. Experiments provide an opportunity for learning, testing each subject, and then, making predictions as to what they will do on individual items.
