Threat appeals and persuasion: Seeking and finding the elusive curvilinear effect

One of the earliest theories of threat appeals, the drive model, predicts that the fear–persuasion relationship is characterized by an inverted U-curve. Evidence of such a curve has proven elusive, probably because, in reality, there are two curvilinear hypotheses, not one. Existing primary studies and meta-analyses have tested the between-persons version of the inverted U-curve and found it wanting. In contrast, this study reports on a within-persons version and finds clear support for it. A peaked curve predicted change in intention to use dental floss, whereas a flatter curve did not. Partitioning the sample on individual differences in the behavioral inhibition system produced curves that did not differ in their ability to predict persuasion. Because no contemporary theory of threat appeals can predict these curvilinear findings, researchers should reconsider the utility of drive theory.