Curso Multigrado

Red light cameras may reduce accidents at intersections in two ways (MacLean, 1985). The first is specific deterrence: drivers who are detected and penalised may be deterred from running red lights in the future. The second is general deterrence: drivers are deterred from running red lights because they believe that if they do, they will be photographed and punished.

It is likely that a large proportion of the accident reductions resulting from red light cameras are due primarily to site specific deterrence effects. Perceived apprehension risk increases as drivers become aware of the increase in the actual risk of apprehension, due to publicity and the visible presence of the red light camera hardware. In the Australian States of Victoria and New South Wales, the site specific effect has been maximised in a number of ways:

(a) by the use of warning signs on all the approaches to a treated intersection, although the camera only operates on one traffic stream;

(b) by making the camera hardware installations (camera housing and flash unit) clearly visible to motorists;

(d) by using widespread publicity to increase awareness of the red light camera program.

Use of warning signs

One feature of the Victorian, New South Wales and South Australian red light camera sites are that they are clearly signposted with the message "RED LIGHT CAMERAS AHEAD". South et al (1989) stated that "it was clear that the maximum deterrent effect would only occur if the presence of the devices was signalised in some way" and concluded that warning signs, as visible symbols of enforcement, were likely to provide the greatest deterrent effect to red light running behaviour.

This notion was supported by Hillier et al (1993) who found that the presence of the red light camera hardware (signposting and camera housing / flash unit) appeared to be effective at reducing right angle and right turn against accidents, even when seldom used as active red light camera sites. It appears that drivers see the installed hardware and are reminded of the possible risk of detection even when they know that the site is not always actively enforced.

The benefits of using site specific warning signs at the approaches to red light camera intersections has recently been questioned. Bodinnar (1993) has argued that, after a suitable time period, drivers learn that only sign-posted intersections present the possibility of red light camera detection and hence modify their behaviour at those sites but are less cautious and law abiding at other non-camera site intersections. This finding is supported by Chin (1989) who reported a significant reduction in red light running at camera sites but no such reduction at non-camera sites.

Bodinnar (1993) suggested that generalised red light camera sign posting should be used as opposed to site specific sign posting as a possible means of addressing this problem. This sign posting practice would still ensure that driver awareness of red light camera operations was maintained, but by not highlighting the precise location of the camera sites, may actually lead to drivers modifying their behaviour at all signalised intersections. A contrary argument has been put forward by Schnerring (1993). He suggested that, since red light camera sites are selected on the basis of accident history, the use of generalised sign posting may reduce the deterrent effect of site specific sign posting and increase the number of accidents at potentially dangerous intersections.

No research evidence was found relating to the deterrent or crash reduction benefits of generalised as opposed to site specific sign posting. However, an evaluation study may soon be undertaken in New South Wales, Australia (Lane, 1993) to examine the benefits of both of these sign posting approaches. This study will involve the use and evaluation of each particular method in one of two similar regional cities.

Camera rotation at sites

The rotation of a small number of cameras through a large number of sites has also been shown to increase general deterrence by maintaining a high level of perceived detection risk at each possible camera site (South et al, 1989). In order to produce the most cost effective coverage of red light intersections whilst maintaining a high level of general deterrence it is recommended (Lawson, 1991) that the ratio of sites to actual cameras not exceed 6 to 1.

South et al (1989) reported that when red light cameras were first introduced in Victoria, Australia there was no formal strategy for rotating cameras among treatment sites. Initially cameras were left at sites until the offence rate started to decline. Once sufficient data on particular sites had been collected cameras were rotated on the basis of violation rates and the need to maintain the level of general deterrence. In Australia it is common practice to rotate cameras among sites every 7 to 10 days. Exceptions to this rule include high red light violation sites, which may require continuous camera surveillance and low violation sites, which require only minimal surveillance.

Use of publicity

Another reported means of increasing the general deterrence effect of red light cameras is by using the media to raise community awareness of red light running and red light camera operations. MacLean (1985) detailed a study undertaken in Victoria, Australia which highlights the effectiveness of the media in modifying driver behaviour at intersections. In 1981 a red light camera, located at a major intersection, was trialed for a three month period. In an initial period, when drivers were not aware of its presence, there were approximately 300 red light violations detected per week. As a result of an intensive period of media publicity this rate dropped to 20 per week. At the end of the trial period the camera was replaced by a dummy unit and this had the effect of ensuring that offence rates continued to remain at a low level. It appeared that the media attention given to this one particular intersection created a lasting deterrent effect.

Thompson et al (1989), reported on the use of red light cameras by the Nottingham County Council in the United Kingdom, and concluded that the reduction in red light running offences was greatest during the period of extensive publicity just after the cameras were officially switched on. Similar publicity campaigns were launched in conjunction with the introduction of red light camera programs in Australia. It is generally accepted that these campaigns had a significant impact on driver behaviour by increasing driver awareness, and hence the general deterrence, of red light camera operations.

Flash unit operations

One final means of increasing general deterrence has been reported by Wray (1985) and involves the deployment of the red light camera flash unit at non-active camera sites. Red light camera flash units can be activated, when a red light runner is detected, independent of the actual camera units themselves. The general policy in most jurisdictions is not to deploy the flash unit when a camera is not present as it may reinforce red light running behaviour when no traffic infringement notification is received by the offending driver. Wray (1985) has argued against this proposition. He suggested that the non-deployment of the camera flash unit at non-active sites may also reinforce red light running behaviour and actually reduce the level of general deterrence. This occurs because the red light runner, and all other drivers present at the intersection, realise, due to the non-deployment of the flash unit, that a red light camera is not operating, and hence no infringement will be recorded. Wray (1985) concluded that the deployment of the flash unit can, at a relatively low cost, increase the general deterrence of red light cameras.

The use of non-detectable, infra-red flash units has also been suggested as a means of increasing the general deterrence effect of red light cameras. When drivers do not see the visible flash deployment they assume that an offence has not been recorded. If the public

are informed about the use of non-visible flash units then, the fact that no flash was observed, would not lead offenders to assume that an infringement had not been recorded. The available evidence indicates that both visible and non-visible flash deployment options have the potential to increase deterrence. However, studies that have examined the visibility of enforcement (Sanderson & Cameron, 1982; Armour, 1984) suggest that a visible flash deployment at a red light camera site may have a greater impact on deterrence due to physical message it conveys than that created with a non-visible infra-red flash deployment.