A Cognitive Framework for Integrated Embedded Training and Decision Aiding

Abstract

This research has developed a theoretically-based cognitive model and design framework for Integrated Decision Aiding/Training Embedded Systems (IDATES). Based on a review of empirical studies of noviceexpert differences and of theoretical and computational models of skill acquisition, we defined a three-stage cognitive hierarchy model as the basis for our IDATES framework. The levels of novice, intermediate, and expert are discrete stages which differ along two primary dimensions: problem representation and problem-solving procedure. Both decision aiding and training must be targeted to the problem representation and cognitive processes of the user/trainee. Thus, there must be three levels of decision aiding targeted to novice, intermediate, and expert decision makers. Furthermore, there are two types of training: incremental training to improve performance within each of the three expertise levels, and representational training to elicit a jump to the next higher level of problem representation. Two implications arise from the IDATES cognitive model. First, integrated cognitive/behavioral task analyses are able to drive both the embedded training requirements and the decision aiding requirements, although the three skill levels must be separately addressed. Second, a single integrated architecture can underlie all the decision aiding and embedded training components of a given IDATES application.