A series of experiments studied how people judge the frequency of death from various causes. The judgments exhibited a highly consistent but systematically biased subjective scale of frequency. Two kinds of bias were identified: (a) a tendency to overestimate small frequencies and underestimate larger ones, and (b) a tendency to exaggerate the frequency of some specific causes and to underestimate the frequency of others, at any given level of objective frequency. These biases were traced to a number of possible sources, including disproportionate exposure, memorability, or imaginability of various events. Subjects were unable to correct for these sources of bias when specifically instructed to avoid them. Comparisons with previous laboratory studies are discussed, along with methods for improving frequency judgments and the implications of the present findings for the management of societal hazards.How well can people estimate the fre-how small a difference in frequency can be quencies of the lethal events they may en-reliably detected? Do people have a concounter in life (e.g., accidents, diseases, sistent internal scale of frequency for such homicides, suicides, etc.) ? More specifically, events? What factors, besides actual frequency, influence people's judgments?The answers to these questions may have to society. Citizens must Defense and was monitored by the Office of Naval assess rlsks accurately in order to mobilize Research under Contracts N00014-76-C-0074 and society's resources effectively for reducing N00074-78-C-0100 (ARPA Order Nos. 3052 and hazards and treating their victims. Official ,5469) under subcontract to Oregon Research In-reco nition of the irn p 0r t a nce of valid risk . , ,.,<<., • ^ » 0722 to Perceptronics, Inc. from Decisions and assessments is found m the vital statistics Designs, Inc. that are carefully tabulated and periodically We would like to thank Nancy Collins and reported to the public (see Figure 1 ). There Peggy Roecker for extraordinary diligence and ; Sj i lowev er, no guarantee that these statispatience in typing and data analysis. We are also .

The Department of Energy's program for disposing of high-level radioactive wastes has been impeded by overwhelming political opposition fueled by public perceptions of risk. Analysis of these perceptions shows them to be deeply rooted in images of fear and dread that have been present since the discovery of radioactivity. The development and use of nuclear weapons linked these images to reality and the mishandling of radioactive wastes from the nation's military weapons facilities has contributed toward creating a profound state of distrust that cannot be erased quickly or easily. Postponing the permanent repository and employing dry-cask storage of wastes on site would provide the time necessary for difficult social and political issues to be resolved.

This study investigates the potential impacts ofthe proposed nuclear waste repository at Yucca Mountain, Nevada, upon tourism, retirement and job-related migration, and business development in Las Vegas and the state. Adverse impacts may be expected to result from perceptions of risk, stigmatization, and socially amplified reactions to "unfortunate events" associated with the repository (major and minor accidents, discoveries of radiation releases, evidence of mismanagement, attempts to sabotage or disrupt the facility, etc.). The conceptual underpinnings of risk perception, stigmatization, and social amplification are discussed and empirical data are presented to demonstrate how nuclear images associated with Las Vegas and the State of Nevada might trigger adverse economic effects. The possibility that intense negative imagery associated with the repository may cause significant harm to Nevada's economy can no longer be ignored by serious attempts to assess the risks and impacts ofthis unique facility. The behavioral processes described here appear relevant as well to the social impact assessment of any proposed facility that produces, uses, transports, or disposes ofhazardous materials.

No abstract