Warning timing and how drivers with and without forward collision warning (FCW) systems react when distracted at the moment a sta- tionary vehicle is revealed directly ahead were investigated. The study was conducted using the Iowa Driving Simulator (IDS). The IDS was equipped with an FCW system that provided auditory warnings based on two warning criteria. A total of 30 subjects were split across three conditions—a baseline of 10 subjects (no warning display), and two warning conditions (early and late) with 10 subjects each. The two warn- ing conditions differed by the duration of an a priori driver reaction com- ponent (1.5 and 1.0 s) in the warning algorithm. Drivers’ collision avoid- ance performance in the two warning conditions was compared with that in the baseline condition. Results indicated that the early warning condition showed significantly shorter accelerator release reaction times, fewer crashes, and less severe crashes than both the baseline condition and the late warning condition. The results indicate that the timing of a warning is important in the design of collision warning systems. Front-to-rear-end crashes involving two or more vehicles currently represent approximately one-fourth of all collisions. Specifically, the National Safety Council reported that there were approximately 11.3 million motor vehicle crashes in 1996 (1), of which 2.7 million were rear-end crashes (about 23.8 percent of the total). According to the General Estimates System and the Fatal Analysis Reporting System, in 1992 there were approximately 1.4 million police- reported (PR) rear-end crashes. Rear-end collisions constituted approximately 23 percent of all PR crashes but only about 4.7 per- cent of all fatalities. While many injuries and fatalities are caused by rear-end crashes, such crashes also cause approximately 157 million vehicle-hours of delay annually, which is approximately one-third of all crash-caused delays. Rear-end collisions can be placed into two main categories: status of the lead vehicle when the collision occurs, and cause of the crash. In 69.7% of all rear-end crashes, the lead vehicle is stopped; in the remaining 30.3% of crashes, the lead vehicle is moving at the time of collision (2). Regardless of the status of the lead vehicle at the time of collision, driver inattention is a major cause of this type of crash. Knipling et al. (3) estimated that inattention accounts for 64% of all PR rear-end crashes and that inattention associated with following a preceding vehicle too closely represents the cause of 14% of rear- end crashes. Although all drivers experience some level of inatten- tion while driving (e.g., talking to passengers, daydreaming, adjust- ing in-vehicle controls, and extra-vehicle distractions), inattention during critical situations may mean drivers will not be able to re- spond quickly enough to avoid a collision. These statistics taken together indicate that crash situations where the lead vehicle is stopped and the driver is not attending to the roadway are an important focus of study. Because rear-end crashes account for such a large percentage of automobile collisions, and because inattention is the most frequent cause of these crashes, there has been considerable research into the possibility of alerting inattentive drivers to potential collision situ- ations (3–9). Several strategies exist to aid the driver in avoiding col- lisions. These strategies differ in terms of degree of intervention. They vary from alerts that suggest subtle speed adjustments to those that initiate automatic emergency braking. Warning the driver about imminent collisions is a promising method for mitigating rear-end collisions (4, 6, 7, 10–15). With this type of system, the driver is warned when a situation is detected that requires immediate re- sponse to avoid a collision. The algorithm that triggers the imminent warning is a critical element in providing meaningful alerts. The algorithm must provide ample warning in dangerous situa- tions without being a nuisance to the driver. To achieve this, the algorithm must trigger the warning far enough in advance that the driver has time to react and avoid striking another vehicle, but not trigger the warning in situations that do not pose a hazard. The prob- lem is that the earlier a warning is provided, the less certain it is that the situation will actually require the driver to act to avoid a colli- sion. Thus, the earlier a warning occurs, the greater the chance that it may be interpreted as a nuisance alarm and therefore desensitize the driver to future system warnings (16 ). A warning strategy that can avoid nuisance alarms may provide significant benefit to drivers. Even a system that could provide a modest decrease in overall reac- tion time of 0.5 s could reduce rear-end crashes by 62% (15). It is clear that a careful implementation of an appropriate forward colli- sion warning (FCW) strategy could reduce the number and severity of rear-end crashes. The primary objective of this study was to investigate whether drivers operating vehicles equipped with an FCW system would, when a stationary lead vehicle was revealed, exhibit enhanced front- to-rear-end collision avoidance behavior relative to those driving without such a system. A secondary objective of this study was to explore whether the timing of the warning affects driver reaction and performance. If an FCW warning system could provide drivers with information to help them avoid collisions, drivers operating a vehicle equipped with such a system would experience fewer Effect of Warning Timing on Collision Avoidance Behavior in a Stationary Lead Vehicle Scenario Daniel V. McGehee, Timothy L. Brown, John D. Lee, and Terry B. Wilson D. V. McGehee, Human Factors and Vehicle Safety Research Program, Univer- sity of Iowa Public Policy Center, 227 South Quad, Iowa City, IA 52242-1192. T. L. Brown, Human Factors Laboratory, NADS & Simulation Center, University of Iowa, 2401 Oakdale Blvd., Iowa City, IA 52242-5003. J. D. Lee, Cognitive Sys- tems Laboratory, Department of Mechanical and Industrial Engineering, Univer- sity of Iowa, Iowa City, IA 52242. T. B. Wilson, Sensor Technologies and Systems, Inc., 7655 East Redfield Road, Suite 10, Scottsdale, AZ 85260. Transportation Research Record 1803 ■ 1 Paper No. 02-3746