Abstract
This paper argues that if the federal government is to successfully protect the public from the adverse effects of environmental noise, its policies will need to be informed by a scientific understanding of the psychological and social factors that determine when noise results in annoyance and when noise may affect health as an environmental stressor. The overreliance of federal agencies on mathematical modeling of average group responses to physical noise levels is discussed as oversimplifying and limiting the understanding of noise effects in crucial ways. The development of a more sophisticated information base is related to policy needs, such as the need to make accurate predictions about the annoyance of particular communities, the need to understand relationships between public participation in noise abatement efforts and annoyance, and the need to identify populations that may be susceptible to stress-related health effects.