Collaborative Research: Understanding Why Vigilance Declines Over Time

Tasks such as baggage screening, military surveillance, and long-haul driving require people to remain focused and attentive over long periods of time. But sustaining attention is difficult and people often start making errors within a half hour or less of starting a monotonous task. This project combines behavioral experiments with statistical modeling in order to determine the kinds of errors people make when the task requires sustained attention and to understand the causes of the errors. Experiments in a high-fidelity driving simulator examine the ways that automated vehicle control affects drivers’ sustained attention. Other behavioral experiments examine factors that reduce people's ability to remain attentive. The project trains graduate and undergraduate students in experimental psychology, cognitive modeling, and human factors engineering. Findings help develop techniques for improving human performance in sustained attention tasks, like providing opportunities for rest and recovery or reducing opportunities for distraction. Monitoring and surveillance tasks are challenging, and operators’ performance drops over the course of 1/2 hour or less. This decline is termed the "vigilance decrement." Despite many years of research and the clear importance of this topic, the processes underlying the vigilance decrement remain the subject of debate. This project uses novel experimental and analytic methods to identify the nature and causes of vigilance failures. The central goal is to isolate the roles of response bias, internal processing noise, and attentional lapses on the vigilance decrement. To do this, the project analyzes psychometric functions, curves that display signal detection rates task as a function of signal intensity, and are described by parameters that directly correspond to bias, noise, and lapse rate. The project examines how the form of the vigilance decrement varies with task characteristics that are believed to influence sustained attention, including task difficulty, mental rest, stimulus information format, and dual-task load. Performance is examined in laboratory tasks and in a simulated driving task. The project delivers empirical data to resolve longstanding debates on the mechanisms of the vigilance decrement, and provides experimental methods and analytic tools to enable future vigilance research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.