We describe a dynamic study of the effect of temporal adaptation on the detection threshold for an object (disk or gratings) in continuously variable luminous surroundings. Two types of visibility threshold were determined: dynamic thresholds, in a nonadapting situation (temporal effect); static thresholds, in conditions of steady adaptation. Two different profiles simulating variations in the luminous surroundings were tested. A transitory increase in the curve of relative threshold versus time was observed. The overshoot amplitude and the position of its maximum in time vary with the test and the luminous profiles selected. The effect of temporal adaptation, evaluated from the overshoot parameters, could play a significant role in practice and lead lighting engineers to enhance luminous levels in the first few hundred meters inside the tunnel entrance.
IntroductionHighway engineers and road safety experts are concerned at the fall in visibility which occurs at a tunnel entrance. Many investigations in either simulated or real surroundings have attempted to resolve this problem('-').It is well known that the visibility of an object is determined by the retinal illuminance distribution in the fovea. In a steady but complex luminous surrounding, image contrast on the retina is reduced by the luminance veil which results from indirect stimuli eccentric to the fovea, which is essentially due to intraocular light scattering(6). In the access zone of a tunnel, and in daytime, the modifications of the surrounding luminance level lead to a wide range of changes in foveal sensitivity depending on both spatial and temporal adaptation of the driver's retina. The resulting fall in visibility has been measured in laboratory simulations, but the relative importance of the two factors could not be specified(7,8) .The present experiment is a dynamic study of the temporal effect of adaptation on the detection threshold of an object in continuously variable luminous surroundings which simulate the access zone of a tunnel. The changes of mean luminance in the visual field when the driver approaches a tunnel entrance are described schematically in Figure 1 -The state of adaptation of the fovea is not significantly influenced by the dark entrance (Zone 1) until the so-called 'adaptation point' P is reached(&dquo;. Up to this point the drivers eyes to 10 the mean 1~lminance ilziue in the open field outside the tunnel; the retinal illuminance decreases because of the progressive reduction of the veiling luminance (Zone 2). At the tunnel entrance (point E) the surrounding luminance level falls abruptly to the luminance level of the road alone (Zone 3). In this zone, as far as adaptation of the eye is concerned, we assume in our experiment that three periods can be roughly discerned after entering the tunnel (Figure 1 (b)). In the first period, the lack of the temporal component of adaptation can be disregarded in comparison to the lack of the spatial one, and the driver perceives an improvement in his visibility due to a noticeable dec...