Cognitive Workload Associated with Different Conceptual Modeling Approaches in Information Systems

Abstract

Conceptual models visually represent entities and relationships between them in an information system. Effective conceptual models should be simple while communicating sufficient information. This trade-off between model complexity and clarity is crucial to prevent failure of information system development. Past studies have found that more expressive models lead to higher performance on tasks measuring a user’s deep understanding of the model and attributed this to lower experience of cognitive workload associated with these models. This study examined this hypothesis by measuring users’ EEG brain activity while they completed a task with different conceptual models. 30 participants were divided into two groups: One group used a low ontologically expressive model (LOEM), and the other group used a high ontologically expressive model (HOEM). Cognitive workload during the task was quantified using EEG Engagement Index, which is a ratio of brain activity power in beta as opposed to the sum of alpha and theta frequency bands. No significant difference in cognitive workload was found between the LOEM and HOEM groups indicating equal amounts of cognitive processing required for understanding of both models. The main contribution of this study is the introduction of neurophysiological measures as an objective quantification of cognitive workload in the field of conceptual modeling and information systems.