As the UK progresses towards wider deployment of partially automated and assisted driving systems, Driver Monitoring Systems (DMS) are becoming a central safety feature. Using in-vehicle cameras to detect signs of distraction and fatigue, these systems are designed to ensure that drivers remain capable of responding appropriately in safety-critical situations.

Most commercially available DMS are developed and calibrated using data from the general driving population. However, this research provides evidence that the eye movements associated with distraction and fatigue differ significantly between regular drivers and drivers with emergency response training, including police and ambulance personnel. Indicators that signal elevated risk in the general population do not necessarily reflect reduced performance in experienced, highly trained drivers.

This has important implications for emergency service fleets operating under high cognitive demand and time pressure. Despite their recognised safety benefits, standard DMS may generate unnecessary alerts or interventions, increasing workload and potentially undermining trust in safety technologies. At the same time, this represents a significant opportunity. By recognising differences in driver expertise, DMS can be designed to better support professional users, addressing their specific needs.

This presentation will outline the evidence supporting these population differences and raise awareness of the need for further investigation to better adapt DMS, improving the management of fatigue and distraction risks in emergency fleets. Recognising driver expertise as a factor in system calibration is an important consideration for future fleet implementation and policy discussions.

Rafael Goncalves, Ph.D. Researcher, University of Leeds

Rafael Goncalves is a Ph.D. researcher at the University of Leeds and a member of the Human Factors and Safety Research team. His research focuses on vehicle automation and on how driver state can be predicted using gaze behaviour. He has experience in driving simulator studies involving automated vehicles and has collaborated on major European projects in the field.

He has also led privately funded research with the company Seeing Machines, investigating methods for readiness estimation for driver monitoring systems. In addition, he was part of an ISO working group on the standardisation of driver readiness conceptualisation and assessment.