Managing Instructional Material

The step from reference works to instructional material is a big one. It does not have to be so, but when a hypertext system is developed for instructional use there is always the wish to include the following: Tests. In many cases it is expected of an instructional system that it is able to test the users on their knowledge of the subject matter. The problem with this is that in many cases simple multiple choice questions, which can be easily handled by a computer, are not enough. Another possibility is to use blanks exercises: The student has to fill in the correct word(s) on open places in the text. Questions that require an open answer are still too difficult for computers to comprehend. In this cases the teacher or instructor will have to evaluate the answers. Coaching or navigating. The user has to be coached to navigate through the material along one of the paths selected by the instructional designers. Depending on a set of criteria, such as results of tests, the user either has to redo one or more of the previous items, or can continue with a new item which covers something extra, or may skip a number of items. The use of Petri nets is one of the possibilities to overcome the typical problems that arise while authoring ‘programmed instruction’. ‘Skill learning’ based upon evidence of having taken the prerequisite steps (detected in a task analysis) makes it quite possible to sequence a learning dialogue. However, planning ‘knowledge acquisition’ creates a notorious problem for the instructional designer, as it is the person’s prior knowledge that defines which concepts are adequate to start with. Giving the student the navigation job still leaves the need to orient him/her in the conceptual dependencies as experts see them. These expert assertions might be definable in so-called Petri networks. They will prompt the student which way to go, but always give the student the authority to overrule this advice.