Resilience of super users’ mental models of enterprise-wide systems

Research on information system training has focused primarily on methods, while neglecting the effects of interruption intervals between training and system implementation. This empirical study examines the resilience of accurate mental models of an enterprise-wide system in a large health care facility. Accurate mental models were shown to withstand the passage of time and resulted in superior field performance following a crash conversion. Consistent with prior experimental research, certain types of learners were more likely to acquire these sounder mental models. In large organisations where the simultaneous training of users on a complex system is virtually impossible, scheduling is an important facet of the training paradigm. The main practical implications of this study involve the development of: (1) training programmes emphasising performance on far-transfer tasks, (2) training schedules designed to minimise knowledge erosion, and (3) criteria for selecting highly capable super users.