Recent educational models of computer-based interactivity stress the important role of a learner's cognition. It has been suggested that interactive learning tasks carried out in the context of an authentic, problem-based scenario will result in deeper elaborative cognitive processing leading to greater conceptual understanding of the material presented. Research methods that have been used to investigate cognition and learning have traditionally included self-report questionnaires, focus groups, interviews and think-aloud protocols and, more recently in computer-based settings, interaction log file or 'audit trail' analysis. While all of these techniques help researchers understand students' learning processes, all are limited in that they rely either on self-report or behavioural information to speculate about the cognitive activity of users. The use of functional brain imaging techniques has the potential to address this limitation. Drawing on issues encountered during a recent study using Functional Magnetic Resonance Imaging (fMRI), this paper discusses the methodological issues involved in the use of these techniques for exploring interactivity and cognition. Initial results comparing brain activation when exploring an interactive simulation with brain activation when using an equivalent tutorial program, for a single participant, are presented in order to provide information about the feasibility of the proposed methodological approach.