The influence of perceptual 'set' on the detection of motorcyclists using daytime headlights.

Voluntary daytime headlight use by the majority of motorcyclists might endanger those not using lights: it has been suggested that drivers might scan for lights rather than for motorcyclists per se. Two experiments are described that attempted to investigate this issue in the laboratory. Subjects had to decide as rapidly as possible whether or not a motorcyclist was present in each of a series of slides depicting traffic. Experiment 1 showed that headlight-using motorcyclists were more quickly detected than unlit motorcyclists, especially when they were far away. However, repeated exposure to headlight-using motorcyclists significantly delayed detection of an unlit motorcyclist. Experiment 2 showed that this delayed-detection effect occurred when only 60% of the motorcyclists shown were using their headlight. Under laboratory conditions, at least, subjects readily appear to develop a 'set' for responding on the basis of headlight-use, even when this is an unreliable guide to the motorcyclists' presence.