Applying visual attention theory to transportation safety research and design: Evaluation of alternative automobile rear lighting systems

• Visual search paradigm used to assess brake lamp detection speed and accuracy with current automobile rear lighting and when tail (presence) lamps are amber.
• Visual search predicts faster and more accurate performance when targets and distractors are separated by color than on other dimensions (luminance).
• Results are similar to results from previous simulator based studies on this topic, support predictions from visual search, demonstrate the value of using the visual search methodology and provide a direction for future research.

This field experiment takes a novel approach in applying methodologies and theories of visual search to the subject of conspicuity in automobile rear lighting. Traditional rear lighting research has not used the visual search paradigm in experimental design. It is our claim that the visual search design uniquely uncovers visual attention processes operating when drivers search the visual field that current designs fail to capture. This experiment is a validation and extension of previous simulator research on this same topic and demonstrates that detection of red automobile brake lamps will be improved if tail lamps are another color (in this test, amber) rather than the currently mandated red. Results indicate that when drivers miss brake lamp onset in low ambient light, RT and error are reduced in detecting the presence and absence of red brake lamps with multiple lead vehicles when tail lamps are not red compared to current rear lighting which mandates red tail lamps. This performance improvement is attributed to efficient visual processing that automatically segregates tail (amber) and brake (red) lamp colors into distractors and targets respectively.