What is programmed instruction and how is it employed in the tutorials?
The technology of automated interactive programmed instruction carefully employs the differential reinforcement of developing behavior. To be reinforced, a behavior must first be emitted. To some extent we privately “emit” behavior as we passively read and listen. In a sense, we “echo” it when we are already inclined to say it. However, during instruction the strengthening of this emitted behavior is not assured without external confirmation and reinforcement. For example, one can read passively “read” or “listen” while actually thinking about something else for some moments. In doing so we miss the concurrence of concepts being presented. This problem is prevented with automated interactive delivery. Concepts are deliberately paired together during instruction and such concurrence is assured when the user MUST OVERTLY emit required components in order to move forward. Thus, differential reinforcement and confirmation of learning can be precisely arranged in sequential steps. Experimental research has clearly demonstrated that required overt emission of behavior during the process of instruction more firmly establishes the behavior and, further, the behavior more readily generalizes to application elsewhere. See Kritch and Bostow, 1998.
The tutorials provided here are significantly different from most on-line instruction. Those familiar with operant conditioning argue that “learning” is best defined as a change in behavior, not a mental process. Even when one “does math in one’s head” these chains of behavior are almost always first established overtly before they recede to the covert level. Unfortunately, most on-line instruction today fails to employ differential reinforcement of active overt behavior. Instead, much of it is little more than a textbook with links to pages and video. Few creators of the on-line instruction are familiar with operant conditioning techniques–significantly those called priming and prompting. In many cases, on-line instruction does not sequentially build from basic concepts to complex ones. In contrast, well-designed programmed instruction requires the construction of a matrix of concepts, sequential ordering of the concepts, and deliberately placed frames that interrelate the developing concepts. In well programmed instruction some frames contain rules and examples, some examples then rules. Review frames are carefully inserted. Generalization to new situations is carefully structured. And most importantly, trial testing of new programs identifies weak instructional sequences that need improvement. With well programmed instruction user PERFORMANCE is the hinge pin for revision. This recognizes that the student is always “right.” The solution to poor student performance is program revision, not claiming student inadequacy. The ultimate test of good instruction is, of course, whether the learner engages more effectively with the world at large.