Background
The overwhelming majority of health behavior studies have reduced the risk perceptions construct to a rating of hazard probability and a rating of severity. Significant positive associations are often, but not always, found between perceived probability and subsequent adoption of health behaviors. Associations between perceived severity and health behaviors are also found, but less often. Nevertheless, with a small number of exceptions, community interventions designed to encourage healthy behaviors by increasing risk perceptions are usually unsuccessful.
Developments in a number of research domains suggest that the reduction of risk perceptions to probability and severity estimates is probably inadequate. For example, recent research on emotional reactions to threat, experience-based and implicit attitudes, the accuracy and variability of risk judgments, and cognitive representation of probability all provide insights that could enrich our understanding of the nature of perceived risk and the processes by which risk perceptions affect risk and precautionary behaviors.
Goals
This invitation-only workshop will bring together scientists from a variety of perspectives to examine critically the perceived risk construct. The short-term goal is to arrive at a more sophisticated understanding of the various components of perceived risk in order to facilitate more productive research on the relation between risk communication, risk perceptions and health behavior. The longer-term goals are to:
- Facilitate the development and enrichment of health behavior theories, and the translation of these theories into research that can predict behavior.
- Guide health promotion efforts toward more effective interventions.
- Assist applied researchers in determining whether their interventions have succeeded in altering the ways in which hazards are perceived.
Among the many questions that deserve attention are the following:
- How do people think about risk?
- Do they carry risk judgments around with them, or only construct them from mental schemas in response to researchers’ questions?
- Are there "natural" risk categories? Are there better or worse ways to ask about perceptions of probability?
- What are the differences between cognitive judgments and experienced-based perceptions in this domain?
- Do they affect behavior differently?
- How can they be assessed?
- Is there a difference between judging oneself vulnerable and "feeling" vulnerable?
- Do risk-oriented interventions--many directed toward teens--fail because they never make people feel vulnerable?
- Do risk-oriented messages fail with some populations because they make people feel too vulnerable?
- Should fear be seen as part of risk perception?
- How should it be assessed in health behavior research?
- What about "worry," "preoccupation," "vigilance," "concern" and other concepts that seem to have both cognitive and affective qualities?
- How do they differ from risk judgments?
- Do characteristics of beliefs such as uncertainty, salience, or accessibility need to be considered in predicting behavior and in characterizing risk perceptions?
- Is "severity" a catchall for various hazard attributes (e.g., vividness, harm, unfamiliarity, temporal proximity) that should be kept separate?
- What dimensions of a hazard should be considered?
Papers
- Conceptualizing and Measuring Risk Perceptions Workshop
Summary of Discussion Sessions - Conceptualizing and Measuring Risk Perceptions
Neil D. Weinstein - Emotional bases for quitting smoking: Extending the concept of risk to discrete emotional consequences
Joseph N. Cappella - Use of the Implicit Association Test in Risk Assessment
Anthony G. Greenwald - Measuring and Conceptualizing Perceptions of Vulnerability/Likelihood
Paul D. Windschitl - Conceptualizing and Assessing Risk Perceptions: A Self-Regulatory Perspective
Linda D. Cameron, Ph.D. - Risk as Analysis and Risk as Feelings: Some Thoughts about Affect, Reason, Risk, and Rationality
Paul Slovic, et al. - Origins and Functions of Perceptions of Risk (PDF)
Elke U. Weber