This paper examines the effect of colour on reaction times for variations in both luminance and chromatic contrast. Results confirm the idea that reaction times are determined by a cone-opponent mechanism: Reaction times generated in response to S-cone isolating stimuli are the longest, whereas the shortest reaction times are generated by L—M-cone isolating stimuli. In addition, an asymmetry between ON and OFF opponent channels is observed for stimuli modulated on the blue—yellow axis. Reaction time (RT) is influenced by hue and chromatic contrast at isoluminant condition. In the isoluminant condition, RT decreases as luminance contrast increases. At luminance contrasts of approximately 20%, RT approaches an asymptotic value, thus becoming independent of colour for further luminance contrast increases. This asymptotic value is achieved for lower luminance contrasts as chromatic contrast increases.

Visual performance is defined as the speed and accuracy of processing visual information. Existing models of visual performance evaluate illuminated tasks in terms of luminance contrast, retinal illuminance and visual size but do not consider the chromatic properties of the task. In consequence, there is no recommendation for the chromatic contrast needed to reach a high level of visual performance. This paper shows that chromatic information is crucial to visual performance when achromatic information is weak or missing altogether. When the luminance contrast is less than approximately 0.20, some task colours with excitation purities greater than 40% can be used to achieve a level of visual performance close to 90%. For luminance contrasts higher than 0.60, visual performance is determined only by luminance information, and tends to the value for an achromatic stimulus. For luminance contrasts in the range 0.20 to 0.60, both luminance and chromatic information are important for visual performance.