ABSTRACT
This chapter discusses methods for the sequencing of task classes that define simple-to-complex categories of learning tasks. Task classes have also been called equivalence classes (Scandura, 1983), problem sets (White & Frederiksen, 1990), and case types (van Merriënboer, 1997). For each task class, the characteristics or features of the learning tasks that belong to it need to be specified. This both allows already developed learning tasks to be assigned to their appropriate task classes as well as additional tasks to be selected and developed. The progression of task classes-filled with learning tasks-provides a global outline of the training program. The structure of this chapter is as follows. First, *whole-task sequencing*, including simplifying conditions, emphasis manipulation, and knowledge progression methods for defining task classes, is discussed. All of these methods start the training program with a task class containing whole tasks representative for tasks encountered in the real world. Then, the issue of learner support for tasks within the same task class is considered. Whereas learning tasks within the same task class are equivalent in terms of complexity, a high level of support is typically available for the first learning task. This support is gradually and systematically decreased in the course of the following tasks until no support is available for the final learning task or tasks within the task class. Third, *part-task sequencing* is discussed. There are exceptional cases where it may be impossible to find a task class simple enough to start the training. In those cases, it may be necessary to break the complex skill down into meaningfully interrelated clusters of constituent skills (i.e., the ‘parts’) that are subsequently dealt with in the training program. Fourth, on-demand education is discussed. Not all learners need exactly the same support and fading thereof. A structure of task classes with learning tasks with different levels of support and guidance can be used as a task database, from which tasks are selected that best fit the learning needs of individual learners and thus yield individualized learning trajectories. In adaptive learning, the tasks are selected from this database by the teacher or another intelligent agent while in on-demand education, they are selected by the self-directed learner who might receive support and guidance for the selection process from the teacher or another intelligent agent (i.e., second-order scaffolding). The chapter concludes with a brief summary.
