Using perceived risk of a heart attack, we examined the relative importance of perceived risk factors and sociodemographic variables on subjects' perception of heart attack risk, the relationship between perceived and objectively measured heart attack risk, and the effect of health risk appraisal (HRA) feedback on risk perceptions. Data derive from a random sample of 732 Greater Boston, Massachusetts area men and women ages 25-65 years, who participated in a field trial of health risk appraisal instruments. At baseline and approximately two months later, all respondents completed a questionnaire assessing their own health-related behavior, risk factors, and perception of heart attack risk. At baseline, respondents also completed one of four HRA instruments. Physiologic measures of cholesterol, blood pressure, and weight were taken at either baseline or follow-up visit. Results showed that respondents used established risk factors in estimating overall risk; that compared to objective measures of risk, a high percentage of respondents displayed an optimistic bias; and that HRA feedback had some effect on perception of heart attack risk for those at high risk. Implications of these results for health promotion are discussed.