Developing Effective Rule-Based Training Using the Cognitive Work Analysis Framework
Robinson, Thomas Alan
MetadataShow full item record
Cognitive Work Analysis (CWA) is a framework for the analysis of complex systems that involve technological tools and human operators who must operate the systems in sometimes-unexpected situations. CWA consists of five phases that cover analysis of ecological aspects of systems, such as the tools in the system and the environment, and cognitive aspects, related to the human operators in the systems. Human operators often need to be trained to use complex systems, and training research has been conducted related to the ecological aspects of CWA; however, there is a gap in the training research related to the cognitive aspects of CWA. The Skills-Rules-Models (SRM) framework is currently the main method of cognitive analysis for CWA; therefore, to begin to fill the gap in the training literature at the cognitive end of CWA, this dissertation examines the need for training as related to SRM. In this dissertation, the current research related to training and CWA is reviewed and literature on the nature of expertise as related to SRM is examined. From this review, the need for training rules that can be used in unexpected situations, as a means of reducing cognitive demand, is identified. In order to assist operators with developing the knowledge to use such rule-based behaviour, training needs and methods to meet those needs must be identified. The reuse of knowledge in new situations is the essence of training transfer, so ideas from training transfer are used to guide the development of three guidelines for determining rules that might be transferable to new situations. Then methods of developing rules that fit those three guidelines by using information from the Work Domain Analysis (WDA) and Control Task Analysis (CTA) phases of CWA are presented. In addition to methods for identifying training needs, methods for meeting those training needs are required. The review of training transfer also identified the need to use contextual examples in training while still avoiding examples that place too high of a cognitive demand on the operator, possibly reducing learning. Therefore, as a basis for designing training material that imposes a reduced cognitive demand, Cognitive Load Theory (CLT) is reviewed, and methods for reducing cognitive demand are discussed. To demonstrate the methods of identifying and meeting training needs, two examples are presented. First, two sets of training rules are created based on the results of the application of the WDA and CTA phases of CWA for two different work domains: Computer Algebra Systems and the programming language Logo. Then, from these sets of rules, instructional materials are developed using methods based on CLT to manage the cognitive demands of the instructions. Finally, two experiments are presented that test how well operators learn from the instructional materials. The experiments provide support for the effectiveness of applying CLT to the design of instructional materials based on the sets of rules developed. This combined work represents a new framework for identifying and meeting training needs related to the cognitive aspects of CWA.