Knowledge Representation
International Encyclopedia of Ergonomics and Human Factors, Taylor & Francis, Boca Raton, (2006)Abstract
A central design principle of cognitive ergonomics is the compatibility of a technical system with the mental model of the user. To minimize transformation efforts it is useful to provide information in a form that complies as well as possible with the user’s mental model being necessary to cope with the task.
In this context, knowledge representation means the formalization or coding of a mental model, for example utilizing a graphical or textual notation. On the one hand, this can be a knowledge representation on a rather perceptual level, e.g. the representation of relational, spatial, temporal, kinematical, dynamic or visual properties or a physical system organization as a device topology. On the other hand, conceptual and causal models highlighting the explanations of functional coherences are used to represent abstract, schematic knowledge. Since there is a co-evolution of cognitive psychology and artificial intelligence (AI), technical concepts of knowledge representation are used in both areas of application. Several notations and languages for knowledge representation were developed, for structuring information as well as for reasoning in a natural way. In AI, these representations should support machined inference mechanisms; in cognitive psychology they are used to explain human information processing. In cognitive ergonomics they are the design framework of a task-appropriate representation for human information processing. The following section gives an overview to eight main models of knowledge representation used in AI as well as in cognitive psychology and ergonomics. Besides the initially described direct representations propositional representations are considered in the following. The semantics of these propositional representations are depicted by different types of symbols, relations and design modalities to comply with their particular application purposes.