Research

The police lights that make the officer harder to see

Thirty-eight years after Illinois said nobody had evidence for roof-mounted lights, a Virginia Tech researcher measured it properly — with radar, cameras and real traffic. Brighter lighting moved drivers over sooner. It also hid the officer standing beside the car.

What was actually tested

Police vehicles were fitted with different lighting, marking and paint schemes and positioned as though conducting a traffic stop. Cameras and radar recorded how approaching drivers behaved — when they slowed, when they changed lanes, and whether they obeyed the Move Over law that now exists in all fifty states.

The locations were ordinary American roads: Interstate 81, Interstate 66, Highway 460, Route 11 and the Prince William Parkway. Data was collected for at least fifteen minutes at each location or until a hundred vehicles had passed, in both daylight and darkness, across marked and unmarked cars.

Then a second, controlled experiment asked a different question — not whether drivers could see the car, but whether they could see the officer standing outside it.

The good news, and it is substantial

Brighter, more aggressive lighting works on traffic. With the enhanced configuration, drivers began reducing speed roughly 600 metres from the police vehicle — well over a third of a mile of advance warning. Adding retroreflective material to the rear of the car made drivers move over sooner still.

The merge data shows how uneven this is in practice. Measured as the percentage of drivers who had already changed lanes between 500 and 700 metres out, the variation between configurations at a single location is enormous — on Highway 460 the figure ranges from 100.57 percent of average down to 31.20 depending on what was on the roof. On Interstate 66, every configuration sat between 87 and 104 percent. Same equipment, different road, completely different driver behaviour.

The finding that should change practice

Here is where the dissertation earns its place. Having established that the enhanced lighting made the vehicle more visible, Terry ran a controlled human factors experiment to check something nobody had: whether it also made the officer more visible.

It did the opposite.

The enhanced blue system degraded drivers’ ability to detect a police officer standing outside the vehicle. The red-and-blue configuration degraded detection too, but only by three metres — about ten feet.

The mechanism is glare. A brighter light source pointed at an approaching driver does not illuminate the scene; it overwhelms the driver’s eyes, and the human figure beside the car falls into the contrast shadow the lights themselves create. The vehicle becomes more conspicuous and the person becomes less so.

Terry is explicit that this was the point of running the experiment at all — to verify that the experimental schemes would not increase risk to law enforcement despite the first phase indicating the vehicles were more visible. That is a researcher deliberately looking for the way his own promising result might get somebody hurt.

What fixes it costs about forty dollars

The study found the problem could be overcome with retroreflective vests worn by the officers.

That is the whole solution. Not different light bars, not a change in the law, not new vehicles. A vest. The lights make the officer hard to see; putting retroreflective material on the officer makes them visible again.

It is worth sitting with how ordinary that finding is, and how easily it could have gone unmeasured. The first phase of the research produced an unambiguously good result — brighter lights, earlier merges, better Move Over compliance. Any agency reading only that would have fitted the enhanced system and moved on, without ever learning that they had made their officers harder to see.

What the officers themselves said

The second phase put the alternative configurations into actual service with local and state agencies for at least eighteen months, and then asked the people driving them.

They preferred the red-and-blue configuration — the one that cost only three metres of officer detection. Their stated reasons were greater comfort for themselves and less glare to approaching drivers.

The officers, in other words, independently arrived at the configuration the human factors experiment had already identified as the safer compromise. They just described it in terms of their own eyes rather than a detection distance.

And one more result, quietly significant for any agency worried that changing the lights changes the job: the alternative configurations did not affect the rate of citations. The cars looked different and the work came out the same.

Reading this against 1982

The two studies are thirty-eight years apart and they answer different halves of the same question.

Illinois asked whether the light bar protected the car and found nothing. Virginia Tech asked whether it protects the person, and found that the brighter you make it, the worse it gets.

Neither study says get rid of emergency lighting, and neither should be read that way. Terry’s work shows real benefits in how traffic behaves. What it establishes is that more is not automatically better — that past a certain point the lights start working against the thing they are there to protect.

Why this belongs in a police car archive

Because the light bar is the one part of a police car that never gets tested.

The engine, the brakes, the tyres, the suspension and the transmission all go through the Michigan evaluations every autumn, measured to the hundredth of a second and published. The single most visible component on the vehicle — the thing that defines it as a police car at all — gets fitted without anyone measuring what it does.

Twice in forty years somebody has. Both times the answer was more complicated than the assumption, and both times the report went into a government archive and stayed there.