Public lighting improves visibility and provides orientation. It also contributes to the perception of comfort and safety of people outside after dark. At present, highpressure sodium lamps are widely used in street lighting. This is in part due to their high efficacy and relatively long lifetime (≥16 000 hours). Their use, however, comes at the expense of good colour rendering. Recently developed ceramic metal halide lamps provide many of the advantages of HPS in addition to white light and better rendering of colours. In this paper, results of research conducted in three European countries on the effect of lamp spectrum on face recognition and the perception of safety and comfort outdoors are presented. The results consistently show that at comparable illuminances, people perceive areas illuminated with white light to be brighter, safer and more comfortable than the same neighbourhood illuminated with yellowish light.

Virtual reality allows researchers to explore training scenarios that are not feasible or are potentially risky to recreate in the real world. The aim of this research was to examine whether using a tutorial session prior to using the mining simulator could adequately reduce the performance variability and increase the consistency of participant performance metrics. Eighteen participants were randomly assigned to a tutorial or a non-tutorial group. The tutorial group completed a five-minute tutorial that introduced them to the basics of the machine and virtual reality environment. All participants then completed five sessions in the simulator lasting five minutes each. Personality scores were recorded and participants answered questions to test their situational awareness after each session. Performance metrics such as number of collisions and perception response time were recorded by the simulator. A Wilcoxon signed rank test was used to determine at what point a significant difference in performance metrics was apparent across the five sessions. A mixed effects multilevel regression was done to evaluate the change in variability across time. There were no significant correlations between the personality questionnaire scores and the number of collisions or the perception response time. Both groups demonstrated high standard deviation scores for collisions and perception response time, but the tutorial group had decreasing variability across time. Both groups began to exhibit more consistent scores in the simulator after 10 min of use. Situational awareness questions require some refinement prior to further testing.