Characterisation of mental models in a virtual reality-based multitasking scenario using measures of situation awareness

The objective of this study was to assess an empirical approach to the characterisation of mental models in a multitasking scenario. A virtual reality task was used, in which participants were required to attend to and detect multiple types of perceptual events occurring randomly in time while carrying on steady physical activity (walking on a treadmill). Based on a cognitive task analysis, different forms of mental models for performing the task were hypothesised. Responses to situation awareness (SA) probes delivered during experiment trials, mental workload ratings and task performance measures were collected and compared with response patterns expected for specific mental model types. Results demonstrated utility of categorical SA responses for identifying different mental model types. However, in the multitasking scenario, the process of developing ‘good’ SA was significantly influenced by the physical and cognitive task demands and the progressive development of accurate isomorphic mental models of event distributions appeared to be restricted by this. Possible applications of the results of this study include the development of training programmes for mobile, visual inspection tasks, such as airport roving security patrols, towards accurate mental model development and promoting detection of critical events.