Improving End-User Proficiency: Effects of Conceptual Training and Nature of Interaction

Abstract

Users of information technology form mental models that reflect' their understanding and knowledge of an information system. These models affect the proficiency with which they use these systems. In this paper, we draw upon assimilation theory of learning to propose and test a two-stage model of mental model development. We examined the effects of two types of training method, namely conceptual model and procedural, and two levels of nature of interaction, namely novel and simple tasks, on end-users' proficiency in forming accurate mental models of an electronic mail system. Our results indicate that the actual mental models of the system formed by the users predict learning success instead of the type of training provided. Subjects who formed mental models that were conceptual in nature performed significantly better than those who formed mental models that were procedural. Main effects for nature of interaction were not significant. However, we observed a significant interaction effect between the models formed by the users and the nature of their interaction with the system. Our findings suggest that end-user performance is enhanced through training methods that provide good conceptual models but only if users form conceptual mental models and retain them.