|need to engage in KBB. It is important to support workers under these situations because analytical reasoning is an effortful and error-prone cognitive activity. Because KBB involves reasoning based on a mental model, it would be useful for the interface to provide an externalized model so that workers do not have to encode, store, and retrieve all of this information mentally. The AH provides a useful basis for achieving this goal because it is a psychologically useful way of representing how the work domain actually functions (see chap. 7). Thus, building a visualization of the AH representation of the work domain into the interface should provide an effective basis for supporting KBB.|
In conclusion, without an overarching framework, it would have been difficult to go from the mass of results in the literature to practical design insights. For one, the relevant findings are spread across several disciplines including human factors, HCI, management science, psychology, and systems engineering. Consequently, the studies use different terms and concepts. This makes it difficult to find regularities in the literature. Furthermore, the pertinent studies were not all originally designed with the purpose of interface design in mind. As a result, it is not very easy to extract reliable design implications. Our example shows how the SRK taxonomy can help us overcome these obstacles. The concepts in the taxonomy can be used to integrate a diverse set of research findings in a way that allows us to derive practical implications for design.
The examples in this section should have shown you the unique value added by the SRK taxonomy to both basic research and systems design. Although it provides only a qualitative framework, its conceptual distinctions provide a very powerful set of "glasses" with which to view the world. This frame of reference can allow us to clarify and enrich existing theoretical claims, to reconcile seemingly competing theoretical accounts, to integrate a diverse set of empirical research findings, and to derive implications for design. Reason ( 1990) provided another example showing how the SRK taxonomy can be used in these ways in the particular area of human error. Even though the contributions of the taxonomy may seem modest initially, it provides us with a coherent conceptual basis for distinguishing between qualitatively different categories of human performance, thereby helping us gain many valuable insights. As Rasmussen ( 1983) pointed out, "we, like Eddington [ 1958] ichthyologist . . . , will be able to obtain some of the results needed more readily by conceptual analysis before experiments than by data analysis afterwards" (p. 266).
In this section, we show how the SRK taxonomy can be used to conduct a worker competencies analysis, as a part of CWA. As in chapters 7 to 9, we use the DURESS II process control microworld as a case study.