ABSTRACT
To design instruction that can predictably improve learning, it is important to specify the learning theory on which the instructional methods and strategies are based. This principle is fundamental to instructional design (ID) theories. Here are some examples of how theories of learning help in defining instructional design:
Thorndike's (1913) initial application of behavioral theory led to his recommendation that classroom teaching methods be based on the laws of effect and exercise.
Pressey's (1926) teaching machine was an extension of Thorndike's learning theory. Pressey's mechanical device, which used a keyboard, presented a series of multiple-choice questions and required the student to respond by pressing the appropriate key. The device also recorded responses to determine whether the student needed more instruction to master the objective. Consequently it made use of a modified form of the law of exercise.
Skinner's (1953, 1954, 1958) instructional design model was based on three basic principles of instruction: (a) small incremental steps of content presentation, (b) overt (active) learner responding, and (c) immediate reinforcement of correct responses. The model further assumed that the student should make only correct responses. In practice, however, this assumption seemed to hold only in those learning situations that were relatively short in duration and that did not have complex associations. Even though Skinner assumed that the small steps could be linked into increasingly complex chunks, his model did not directly show how to design the more complex chunks such that responses would always be correct. Attempts to correct the basic behavioral paradigm are seen in Crowder's contributions.
Crowder's (1960) intrinsic programming allowed the more able learners to branch more quickly through the instruction while providing corrective frames for those who missed a question. And, although Crowder and others did not state an underlying learning theory or empirical evidence to support the branching technique, it clearly is in the behaviorist tradition. It became the design strategy of choice for computer-assisted instruction, especially for drill and practice, tutorial, and testing strategies.
Contemporary cognitive theories of learning are having a similar effect on current ID theories (e.g., Salomon, Perkins, & Globerson, 1991). The concern of current theories with knowledge structures, metacognitive strategies for problem solving, and integration of new and existing knowledge structures by the learner is leading to a number of ID changes, including: development of a new theoretical basis for design of learning environments intended to facilitate exploration and reinforce context (e.g., computer “microworlds” and hypertext); significant expansion of the role of learner control coupled with an enriched dialogue between the learner and the learning environment in order to support individual learner differences, facilitate integration, and enhance motivation (e.g., the “coach” technique used in intelligent tutorial systems); and a redefinition of the structural requirements for simulations and games.
