The Federal Highway Administration (FHWA) and the Pennsylvania Department of Transportation (PennDOT) undertook a cooperative effort to study nighttime driving performance through curves. The first experiment in this effort was a field experiment conducted on a curvilinear rural Pennsylvania roadway. Each night various curves were treated with different combinations of pavement markings, pavement markers, and horizontal signage. The second experiment was a realistic laboratory simulation of the initial field study. In both experiments research participants rated the effectiveness of the roadway delineation on each curve by means of a 7-point scale. An important methodological question was: can such a subjective rating scale be effectively used in a driving simulator to reveal real-world stimulus-response relationships? The mean effectiveness ratings (MERs) were similarly sensitive to improved roadway delineation both in the simulator and in the field. MERs generally increased with increasing luminance/retroflectivity of pavement markings and markers in both testing environments. Luminance trading ratios obtained from the simulator ratings were generally consistent with those reported in previous studies. The subjective rating scale was only partially effective in demonstrating the efficacy of applying wider edge lines or horizontal warning signs. In general, the subjective rating scale proved to be effective both in the simulator and in the field, only the scale tended to produce more variability and weaker overall results in the simulator relative to the field. The simulator may require a larger sample of research participants and more repeated experimental trials to produce equally strong effects.
Abstract