|Date added||June 30, 2016|
|Category||2016 CARSP XXVI Halifax|
|Tags||Research and Evaluation, Session 4B|
|Stream/Volet||Research and Evaluation|
Background/Context: Motorcyclists are often involved in collisions where the driver of a car “looked but failed to see” the motorcycle before turning left across its path. Much of the literature focuses on sensory factors (size, lighting, and colour) that contribute to motorcycle conspicuity, and suggests countermeasures aimed at increasing motorcycle visibility (e.g. fluorescent jackets). However, despite motorcyclists’ efforts to increase their conspicuity, “failure-to-see” collisions are increasing, and the assumption that motorcycles are “invisible” in traffic remains largely unchallenged. Previously, we examined whether motorcycles were less visible than cars in a traffic environment using signal detection theory and change-blindness, and concluded that this assumption was false: All experiments showed higher sensitivity to motorcycles than cars, without any conspicuity-enhancing treatments applied to the motorcycles. We concluded that these “failure-to-see” collisions are due to poor driver decision-making rather than motorcycle conspicuity and proposed active countermeasures to reduce motorcyclist involvement in these types of collisions.
Aims/Objectives: As sensory factors cannot explain the increased sensitivity to motorcycles observed in our experiments, the present work aims to identify and describe the psychological factors that cause motorcycles to stand out in a traffic environment. To this effect, we employ a change-blindness paradigm in which participants report whether they detect a change in a flickering image.
Methods/Target Group: All participants viewed a series of flickering images and indicated whether they detected a change.
Experiment 1: Images involved traffic scenes in which a target object (car, motorcycle, or traffic-unrelated) was removed between flickers.
Experiment 2: Images involved a grid of 20 random distractors in which a target (car, motorcycle, or chair) changed between flickers.
Experiment 3, 4, and 5: Images were identical to Experiment 2, except for the distractors, which were all motorcycles or cars. Experiment 3 held distractors at equal frequency, Experiment 4 varied the frequency of these distractors, and Experiment 5 varied the frequency at which the target was a car or a motorcycle.
Results/Activities: Data collection is ongoing for Experiments 3, 4, and 5. Experiment 1 shows increased sensitivity to motorcycle targets in a traffic environment. Experiment 2 eliminates the traffic context and shows no increased sensitivity to motorcycles. Experiment 3 reinstates the traffic context by presenting only images of vehicles; we expect the return of increased sensitivity to motorcycle targets. Experiments 4 and 5 will determine whether sensitivity to a vehicle type is due to the frequency of its category in the display or to the frequency at which its category is the target.
Discussion/Deliverables: Together, these experiments show that motorcycles are not inconspicuous in a traffic context, at least when compared to cars, and that their high detection rates in that context are due to schematic processing. Experiments 3, 4, and 5 test whether this schematic processing is driven by motorcycle prevalence in the scene or in the observer’s memory.
Conclusions: While somewhat counterintuitive, it seems that motorcycles are readily detected in traffic, as their relative rarity causes them to stand out amongst common vehicles.