A loss of sufficient Situation Awareness (SA) may lead to human errors, possibly resulting in accidents. SA is often conceptually described in terms of operators' correct perception and understanding of a situation. This conception of individual SA has been summarised by saying that an operator should be able to (1) recognise the relevant elements in a situation, (2) understand how these elements are interacting and, on the basis of this understanding, (3) predict the system status into the immediate future (Endsley, 1993). A high degree (1-3) of SA is usually related to good performance.

System information is often presented visually. Therefore it makes good sense to base the SA measure on operators' accessing of visually presented information (Hauland, 1996). Operators communicate through the process interface, i.e. they may infer what the other operator is aware of by watching key parameters in the process interface. Such Areas of Interest (AOI) may be substitutes for questions concerning each process parameter. It should be possible to define AOI's relevant to team SA and measure the line of gaze towards these AOIs by using eye movement tracking. Operators will also communicate v erbally with each other. This communication may reveal how well the operators understand the developing situation. In studies conducted by Risø and the Danish Maritime Institute (Andersen et al., 1996), a significant correlation was found between crews' communication related to future system states and performance in simulated (ship manoeuvring) tasks.

These initial ideas for a continuous measure of (team) SA were tried out in a small pilot study conducted in the reactor control room at Risø. The reactor control room is a real operative environment where the operators have to co-ordinate their tasks to achieve the desired level of safety and efficiency. The two main objectives of this technical pilot study was to (1) gain hands-on experience with the iView eye tracking system (SensoMotoric Instruments) and The Observer Video-Pro (Noldus Information Technology), as well as the specific set-up of this equipment in the field, and (2) to get insights related to the implementation of these suggested measures in analyses of team SA. In specific we looked at the feasibility of: Using head tracking in an operative environment, (accuracy/magnetic interference), defining planes and objects for automatic analyses of EMT and defining procedures for measuring team situation awareness.

References

1. Andersen, H.B.; Sørensen, P.K.; Weber, S.; Sørensen, C. (1996). A study of the performance of captains and crews in a full mission simulator. Risø National Laboratory, Roskilde. Risø-R-916.
2. Endsley, M.R. (1993). Situation awareness in dynamic human decision making: measurement. Proceedings of the 1st International Conference on Situational Awareness in Complex Systems (Orlando, February 1993).
3. Hauland, G. (1996). Building a Methodology for Studying Cognition in Process Control: a Semantic Analysis of Visual and Verbal Behaviour. Post-Graduate Thesis in Psychology, Norwegian University of Science and Technology.

© 2000 Noldus Information Technology b.v.