Cognitive Structural Modelling of Skills for TEL .doc

Cognitive Structural Modelling of Skills for Technology-Enhanced Learning Dietrich Albert, Cord Hockemeyer, Birgit Mayer, and Christina M. Steiner Department of Psychology, Cognitive Science Section University of Graz, Austria |dietrich.albert|cord.hockemeyer|bir.mayer|chr.steiner @uni-graz.at Abstract This paper suggests the combination of cognitive and structural modelling as a formal framework for intelligent learning systems. It takes up the distinction between declarative conceptual and procedural action knowledge in the sense of ACT-R. Additionally, prerequisite relations between skills of a domain are captured according to Competence-based Knowledge Space Theory. Benefits, drawbacks, and TEL applications of the outlined modelling approaches are discussed. Introduction Cognitive and educational psychology provided considerable contributions to foster the development of learning technologies that model and adapt to the knowledge of individuals [1]. Cognitive modelling, such as the ACT-R Adaptive Character of Thought - Rational Theory e.g. [2] , and frameworks for structuring knowledge, such as Knowledge Space Theory KST, e.g. [3] , have been successfully utilised for realising technology enhanced learning TEL systems e.g. [4][5] . Cognitive Modelling: ACT-R Theory The ACT-R [2][6] is a production system theory for specifying how human cognition works. The central assumption is that complex cognition arises from the interplay of declarative conceptual and procedural action knowledge [6]. Declarative knowledge encodes the conscious factual knowledge and is represented in propositional networks. Procedural knowledge is displayed in behaviour and refers to mostly unconscious cognitive skills. It is represented in if – then production rules that operate on the declarative memory. The ACT-R framework has been effectively deployed to create models in domains like learning, memory, and cognitive development [6]. Also TEL systems for teaching