This paper presents lessons learned from building a wide range of Adaptive Instructional Systems (AISs), ultimately bearing on the question of how to characterize the space of potential AISs to advance the cause of standardization and reuse. The AISs we consider support coached practice of complex decision-making skills—e.g., military tactical decision-making, situation assessment, and systems troubleshooting and management. We illustrate forces that affect system design and dimensions along which systems then vary. The relevant forces derive from the AIS’s area of application, the project structure within which it is built, and the customer’s priorities. Factors to consider include (1) The extent to which the target domain is well-defined versus ill-defined; (2) The degree of fidelity required, preferred, and/or available for an exercise simulation environment; (3) The intended roles of automated and human instructors in instructional delivery; and (4) Imperatives for short-term and/or long-term cost containment. The primary dimensions of AIS design we consider in this paper include (1) Exercise Environment; (2) Knowledge Models; (3) Tutor Adaptations; and (4) Supporting Tools. Each of these is further broken down into a set of more detailed concerns. Together, they suggest structures that can inform an ontology of AIS methods and modules.
CITATION STYLE
Domeshek, E., Ramachandran, S., Jensen, R., Ludwig, J., Ong, J., & Stottler, D. (2019). Lessons from Building Diverse Adaptive Instructional Systems (AIS). In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11597 LNCS, pp. 62–75). Springer Verlag. https://doi.org/10.1007/978-3-030-22341-0_6
Mendeley helps you to discover research relevant for your work.