Self-Improvable, Self-Improving, and Self-Improvability Adaptive Instructional System

Abstract

Instructional systems that incorporate approaches and techniques that are geared to accommodate individual differences in students are termed adaptive instructional systems, and they are typically further distinguished as self-improvable, self-improving, or self-improvability. These adaptive instructional systems have long been viewed as common and effective forms of education, dating back as early as the fourth century BC. Producing computer tutors capable of achieving levels of effectiveness that rival human tutors has been an ultimate goal for learning scientists and cognitive psychologists who develop intelligent tutoring systems (ITSs). Research, development, and innovations in learning sciences occurring over the past four decades have made it possible for some of the ITS to perform nearly as well as human tutors. The effectiveness of ITS is continually improving due to ongoing theoretical research and technology advancement;, however, these consist chiefly of archiving new designs and producing version updates. Consequently, ITSs stop improving as soon as the development ceases. As human tutors are best viewed as learning resources to human learners, they in turn are constantly learning to adaptively improve their teaching from their interactions with students. This self-improvability is fostered on these self-improving educational interactions over time. Ideally, advanced ITS should be similar to human tutors and self-improve over time as well. In this chapter, we explore the potential of a particular ITS that can self-improve. Here our focus is on a more general category of learning systems, namely, an adaptive instructional system (AIS). More specifically, we introduce three concepts that allow AIS to self-improve—the self-improvable potential of AIS, self-improving process of AIS, and self-improvability used as a measure of the degree of improvement of a self-improving AIS.