Reading the Road

The actual riskiness of New Zealand state highways is measured and mapped in a partnership between the AA, the Transport Agency, the Ministry of Transport and ACC under KiwiRAP (Kiwi Road Assessment Programme). Video and other data were assessed and each 100m segment of state highway assigned a score based on pre-determined factors known to contribute to injury. These road segments were then merged and given a risk rating between one star (unsafe) to five stars (very safe), based on the average scores.

Engineers use KiwiRAP to prioritise road safety improvements, but drivers may be unaware that one road is not as safe as another; many drive as if all roads were the same. That realisation led the KiwiRAP partners to question if there was a difference between the road risk, as drivers perceive it, and the actual risk. If there was, it might partially explain why drivers don’t take enough precautions in certain places.

Working with the Transport Agency, the AA Research Foundation commissioned the Waikato Traffic and Road Safety (TARS) Research Group and engineering firm Beca to investigate. The research was extremely thorough, putting ordinary New Zealanders in simulators and real-world driving situations and using eye-scanners to see which hazards were perceived and which were missed. It also noted that drivers scored roads for risk reasonably consistently, whether they drove them, watched a video of someone else driving them or just looked at a still image.

The researchers found that, in general, drivers’ main perception of road risk is derived from the potential difficulty of steering it and the obvious consequences
of failing to do so. So, while drivers perceived high workload steering situations — like corners with banks and cliffs — as hazards, or bridges with barriers on each side
as high-risk, they didn’t notice roadside dangers such as trees, poles and ditches lining straight roads. They also had a remarkable amount of faith in other drivers not to
be erratic at intersections, even though experts rate intersections as very risky.

The findings show fairly conclusively that drivers’ perception of risk and steering workload are interrelated. When hazards don’t create any steering workload, drivers don’t notice them. As a study from Canterbury University found, this can be a problem because drivers seem to be more likely to succumb to drowsy-driving crashes an hour after they pass from a high steering workload road, like a twisty hill, to a low workload one, like a long boring straight. But these straights, with their unnoticed poles and ditches, can be deadly.

The research has been hailed as world leading, but fitting it into the current policy preoccupations of Government is not easy. One of the big pushes from the Ministry of Transport, Police, the Transport Agency and professional groups like Trafinz is for speed limits to be reduced to lower the risk of collisions. But, do people reduce their speed when they perceive more hazards?

To shed some light on this specific question, the AA carried out a Member survey.

Although it is not of academic quality, with 1,300 usable responses the survey carries reasonable weight. What we found was that when shown pictures of roads, respondents said they would drive them at speeds close to the speed limits. Even when shown a picture of an autobahn and told there was no speed limit, Members opted for, on average, 106km/h. It took a particularly high steering workload rural road or a suburban street to make most Members say they would drive at slower speeds.

What this shows is that people already have a clear idea of what speed they can drive a road just by looking at it, and speed limit signs don’t change the perception of its riskiness. Reducing speed limits will undoubtedly generate more tickets, but is unlikely to change these perceptions.

Members also have clear ideas on fair enforcement. Shown pictures of open and urban roads and asked at what speed someone should be given a ticket, the response averaged at around 8-10km/h above the speed limit. This was the same for both males and females. However, in a separate survey, when we specifically asked about Police reducing the tolerance before they would start ticketing (from 10km/h over the limit to 4km/h over the limit for the Christmas holiday period), there was just over 50% support. Significantly more women supported the initiative than men. Men, who are more likely to be ticketed, were also more likely to see the move as revenue gathering. Both women and men agreed (81-83%) that the tolerance on passing lanes and motorways should remain at 10km/h.

All of this suggests that there is a considerable opinion gap between road safety professionals and the general public. With the road toll reaching record lows, trying to convince drivers that every journey they make should be slower because an increasingly small minority are being killed or injured is going to be a tough sell.

While there will definitely be places where speed limits do not correspond to innate road risk and limits should be reduced, speed limits by themselves are clearly not the whole answer. Deaths and injuries caused by drowsy-driving, misunderstanding road conditions or misreading traffic will not be cured with speeding tickets. To have a lasting impact on road trauma, we need to close the gap between what drivers recognise as dangerous and what is dangerous. This AA Research Foundation funded research is a vital step on that journey.