Psychological Bulletin Copyright 1997 by the American Psychological Association, Inc. 1997, Vol. 121, No. 1, 3-19 0033-2909/97/$3.00 Shaping Perceptions to Motivate Healthy Behavior: The Role of Message Framing Alexander J. Rothman University of Minnesota, Twin Cities Peter Salovey Yale University Health-relevant communications can be framed in terms of the benefits (gains) or costs (losses) associated with a particular behavior, and the framing of such persuasive messages influences health decision making. Although to ask people to consider a health issue in terms of associated costs is considered an effective way to motivate behavior, empirical findings are inconsistent. In evaluating the effectiveness of framed health messages, investigators must appreciate the context in which health- related decisions are made. The influence of framed information on decision making is contingent on people, first, internalizing the advocated frame and, then, on the degree to which performing a health behavior is perceived as risky. The relative effectiveness of gain-framed or loss-framed appeals depends, in part, on whether a behavior serves an illness-detecting or a health-affirming function. Finally, the authors discuss the cognitive and affective processes that may mediate the influence of framed information on judgment and behavior. To the extent that people are motivated to seek health and avoid illness, healthy behaviors should be easy to promote. The opportunity to obtain a prostate examination, for instance, should be embraced with little hesitation when the costs of missed early detection are made salient. In fact, the impact of appeals that emphasize personal vulnerability is predicated on the assumption that people will adopt an available, effective behavior to reduce the likelihood of experiencing an unwanted outcome (e.g., Gerrard, Gibbons, & Bushman, 1996; Sutton, 1982; Weinstein, 1993; Weinstein, Rothman, & Nicolich, 1996). Although the particular effectiveness of fear- or vulnerability- based appeals has been inconsistent, the broader assumption that behavioral responses reflect the manner in which people conceptualize a health threat appears valid (Clark, 1994; Salo- vey, Rothman, & Rodin, in press; Skelton & Croyle, 1991). Actions are best understood in terms not of the objective features of a health issue but rather of the features that people attribute to the issue (e.g., Baumann & Leventhal, 1985; Cioffi, 1991b; Leventhal, Nerenz, & Steele, 1984; and Meyer, Leventhal, & Guttman, 1985; see Leventhal & Diefenbach, 1991, for a re- view). Given the importance of these perceptions, substantial effort is devoted to shape the public's views on health issues Research was supported in part by National Institute of Mental Health Grant MH10576; American Cancer Society Grants PBR-84, PBR-84A, and PBR-84B; National Cancer Institute Grant P01-CA42101; and Na- tional Science Foundation Grant BNS-9058020. We thank Brian Bedell, Delia Cioffi, Jerusha Dertweiler, Kristi Lemm, Anne Moyer, Duane Wegener, and Janet Zullo for their comments on an earlier draft of this article. Correspondence concerning this article should be addressed either to Alexander J. Rothman, Department of Psychology, University of Minne- sota, Twin Cities. Elliott Hall, 75 East River Road, Minneapolis, Minne- sota 55455, or to Peter Salovey, Department of Psychology, Yale Univer- sity, P.O. Box 208205, New Haven, Connecticut 06520-8205. Electronic mail may be sent via Internet to rothm001 @gold.tc.umn.edu (Rothman) or psalove@yalevm.cis.yale.edu (Salovey). through information campaigns on public transportation, in newspapers and magazines, and on radio and television. The power of these campaigns is revealed in often contentious battles over what information should be presented to the public (e.g., Should information about AIDS emphasize safer sex or total abstinence?). An intervention can render any aspect of a health issue salient. Are there particular advantages to emphasize one set of features over another? Does it matter whether an appeal to promote condoms emphasizes the benefits of protected sex or the costs of unprotected sex? People can be sensitive to whether a behavioral alternative is framed in terms of its associated costs (loss frame ) or benefits (gain frame 1), even when the two frames describe objectively equivalent situations (Tversky & Kahneman, 1981 ). To account for this shift in preferences, prospect theory proposes that people are more willing to accept risks when they evaluate options in terms of associated costs but act to avoid risks when the same options are described in terms of associated benefits (Tversky & Kahneman, 1981). For example, when people choose between two treatment programs framed in terms of the number of lives that will be lost, they risk the possibility of greater losses to avoid a certain loss (e.g., they prefer a program that provides a 33% chance of no patients dying and a 66% chance of all 600 patients dying to an alternative program in which 400 patients are sure to die). When the same programs are described in terms of the number of lives that will be saved, people become more conservative in their preferences. They forego the opportunity for greater gains, in exchange for an alternative that provides a certain gain (e.g., they prefer a pro- gram in which 200 patients are sure to be saved to an alternative program that provides a 33% chance of saving all 600 patients and a 66% chance of saving no one). Note that, although the frame shifts in the two scenarios from lives lost to lives saved, The specific labels loss and gain frame are used throughout this article instead of the more general labels negative and positive frame.