ABSTRACT
This chapter focuses on to increase the power of cognitive theory of learning by imposing more structure on it—that is, by introducing additional hypotheses which make more refined predictions possible. It considers the different ways of extending the partial theory of motivation. The chapter also focuses on the problem of control, or how to manipulate motivation. Useful predictions may be obtained in relatively unstructured situations by modifying the motivation mechanism to allow for probabilistic, rather than deterministic, selections. The problem of motivation theory is formulated as follows: Given a goal situation (S, G), and a finite set of relevant rules which may satisfy (S, G). In hypothesis-testing theories, it is generally assumed that subjects select from among available concept attributes in strictly random fashion. The chapter considers some specific experience-free hypotheses which might reasonably be proposed to account for rule selection.
