This report presents the results of a study in which four different ways of designing `self-explaining roads' were evaluated. Design 1 was a baseline, derived from existing specifications of functional and operational criteria for road design. Design 2 used road surface characteristics (texture and colour), and the dimensioning of the centre lane separation markers, in order to increase the discriminability of road categories. In Design 3 `counter-stream' markings and side markings were applied. Design 4 made use of shoulder marker posts as the most prominent potentially discriminative feature. The four designs were compared in four separate experiments, each focusing on a specific quality index derived from categorizability and learnability criteria. The main conclusion from the experiments is that the use of `counter-stream' markers, in combination with side markings, has very favourable effects on the discriminability of road categories. The finding that estimates of expected maximum safe speeds were higher than intended, on the `lower' road categories, may mean that the behaviour to be displayed is not yet the intended one, even if the different road categories can be perfectly categorized per se. (A)
Abstract