An investigation of the role of vehicle conspicuity in the ‘Looked but failed to see’ error in driving

Accidents involving motorcycles tend to occur at intersections where a car driver
claims to have ‘looked’ but failed to have ‘seen’ them. Motorcycle accidents
accounted for by the ‘looked but failed to see error'(L.B.F.S.) are traditionally
explained in terms of the motorcyclist’s relative lack of conspicuity compared to
cars. After reviewing the literature and assessing the diverse methods available to
investigate the L.B.F.S. error this thesis refines previous experimental methodologies. Evidence is then gathered from a police accident database, and by laboratory and field studies, to support the view that poor physical conspicuity is not the only explanation for motorcycle L.B.F.S. accident.

The theoretical stance taken is that driver expectancies may account for, at least
in part, some of these accidents and that the functioning that human attentional
systems during driving may aid the understanding of this error.

Experimental chapters investigate the amount of time drivers search at junctions and consider if object recognition theories can account for possible detection failures of uncommon vehicles such as motorcycles. These studies are complemented by investigations of driver eye-movements in the laboratory which compare detection performance of novice, experienced and expert drivers.

Results suggest that experienced drivers adapt through time to perform driving as possibly an automated process neglecting uncommon vehicles that may be encountered on the highway, even if they are conspicuous in sensory terms. To illustrate this, the final experimental chapter reports accidents involving conspicuous police vehicles which are hit by drivers who claim that they did not see them.

Tentative conclusions drawn suggest that experience need not mean expertise and that the L.B.F.S. error may be a cost of experience.