The safety effects of (digital) roadside advertising: an overview of the literature

Deliverable 1.1a of the CEDR project ADVERTS
Vlakveld, W.; Helman, S.

Roadside advertising is, by its nature, designed to capture the attention of road users. Because humans have limited attentional resources it is therefore possible that such advertisements could hamper the safe execution of the driving task. The aim of this systematic literature review is to summarise findings from existing research on the impact of roadside advertisements on distraction and road safety (with a specific focus on digital advertisements).

An adapted version of the SEEV (Salience, Effort, Expectancy, Value) framework about visual attention of road users (Horrey, Wickens, & Consalus, 2006) was used to structure the findings. The factor of ‘Luminance’ was added to the SEEV-framework and the existing factor of ‘Value’ was expanded. These changes were made, respectively, to ensure that the issues of glare and billboard content were considered along with salience, the effort required from drivers, and the impact of expectancy.

Fifty studies were identified about the effects of roadside advertisements on crash risk or road user behaviour. Forty-nine of these studies were about drivers and one was about motorcyclists.

Five studies examining the association between crashes and roadside advertisements were found, although the findings do not allow firm conclusions to be drawn. In one study an increase in the number of crashes was found near billboards and in the four other studies no change was detected. However, the literature suffers from at least two limitations. One issue is that existing studies may lack statistical power. Because crashes are rare events and the stretches of road on which a billboard could contribute to a crash are short, the included numbers of crashes in such studies are mostly small, and therefore require many observations to detect any changes. Another issue is research design. Three of the five studies were before-and-after-studies (the crash rate before the placement of billboards and the crash rate after) and in two studies a comparison site was also included. A before-and-after-study with a comparison site is a stronger research design than a before-and-after-study without such a group. There were two before-and-after-studies with what appeared to be sufficient statistical power and a comparison site. In one of them an increase of crashes near digital billboards was found and in the other no difference was found.

Approximately half of drivers do not look at billboards when they pass them. However, drivers look more often and for longer at so called static digital billboards that display various static advertisements in sequence, when compared with traditional static billboards. The moments during which advertisements switch seem to be particularly distracting. Video billboards (digital billboards that display constantly moving images) attract the most attention. Drivers more often look at billboards at street level than at raised level but when they look at billboards at raised level they tend to look for longer. Young novice drivers are inclined to look at static digital billboards and video billboards the most. Although drivers rarely look for more than two seconds at a billboard, they sometimes do, especially when it is a static digital billboard or a video billboard. Given that there is strong evidence that long glances at objects outside the vehicle (and away from the road) increase crash risk substantially, this finding may be a cause for concern.

There is quite strong evidence that lane keeping deteriorates when drivers look at billboards. Billboards do not seem to have a strong impact on speed. In some studies, headways were found to get shorter, in particular when drivers pass static digital billboards and video billboards. This may be dangerous because response latencies are longer when drivers look at billboards, and shorter headways mean that safety margins may be decreased even further. Drivers also more often overlook road signs and tend to forget to signal when they change lanes near billboards. The content of what is displayed on billboards does not seem to influence driving behaviour very much. What is displayed on the advertisements (arousing pictures and texts) influences decision making and driving performance only slightly. However, more research is needed to assess the effect of emotion-laden pictures and text on driving behaviour.

The less demanding the driving task is, the more drivers let their eyes wander to task-irrelevant objects. It seems that most drivers are able to regulate adequately the attentional resources the driving task requires. However, in traffic the task demands can suddenly increase and critical events (e.g. a lead vehicle that suddenly brakes) may not always be properly anticipated. In those circumstances looking at billboards could have detrimental effects on safety.

In most field studies no difference was found in gaze behaviour at illuminated billboards during day and night. This does not mean that such billboards do not have an impact on driving at night. Most jurisdictions regulate the amount of light billboards are allowed to emit. The illuminated billboards drivers passed at night in these field studies most probably were in compliance with these regulations. Illuminated billboards definitely can cause glare. In one simulator study it was found that the more light a digital billboard emitted the longer were driver response latencies to acute road hazards.

Gepubliceerd door
CEDR, Brussels


Dit is een publicatie van SWOV, of waar SWOV een bijdrage aan heeft geleverd.