Spare visual capacity and driver inattention in dynamic car following scenarios

Abstract

Drivers often look away from the forward roadway. However, this does not necessarily mean that they are inattentive, as they might be utilizing their spare visual capacity (SVC). Because of the safety–critical risks associated with inattentive driving, it is imperative to analyze how much visual capacity a driver can afford to use for secondary activities while avoiding the possibility of an accident. This study aimed to define and identify driver inattention in a car following task based on an estimate of the driver’s situational SVC. A mathematical model of SVC in car following was created based on the possible but unlikely worst-case scenario in any situation. The model generates situation-specific marginal values of the minimum time that can lead to a possible collision when a driver looks away from the lead car. A classification algorithm was developed to define and identify inattention based on the model’s estimates. The model reveals that SVC in car following varies significantly from situation to situation, depending on relevant situational, technical, and cognitive factors. A driving simulator study (N = 32) indicated that drivers are often willing to occlude themselves in car following when there is a chance of a rear-end collision and that this behavior becomes more likely with higher speeds. A strong linear association between distance headway and brake response time was also found. Quantifying SVC in driving helps in determining driver inattention against a valid baseline based on what is possible. The proposed modeling approach can be utilized for the development of improved safety guidelines and effective context-sensitive (in)attention monitoring systems.