TICL 2: Knowledge Representation, Associated Theories and Implications for Instructional Systems Dialog on Deep Structures

摘要

Both ITS, on the one hand, and ID based relational models, on the other, have important limitations. ITS's focus in my opinion too much on assumptions about neural components that might be in learner's minds. ID largely minimizes, if not ignores the possibility of adaptive instruction - except as realized in self-directed learning. Self-directed learning may work reasonably well with sophisticated learners - and may not be a bad choice in some subjects at the college level. Even so, experience shows that various forms of assistance - currently called scaffolding - may be essential. What scaffolding? And, when should it be given?rnRepresenting what must be learned as SLT rules not only makes learning highly efficient, but it also makes system development more systematic and reliable. Defining pedagogy independently of content semantics has the added advantage of dramatically reducing development costs (cf. Scandura, 2005, Foshay & Preese, 2005).rnThis audience includes very few who are intimately familiar with SLT, with many more production system (and to a lesser extent and CBM) proponents, plus a large cadre of instructional designers. Accordingly, I look forward to this opportunity to hear and respond to your thought provoking questions and comments.