Abstract
The present study tested the hypothesis that the relationship between fear arousal and persuasion is a curvilinear one. High-, moderate-, and low-fear messages were presented to over 900 students in 6 public schools. In addition to paper-and-pencil measures, a behavioral measure of toothbrushing was used. In general, the results did not support the curvilinear hypothesis but did tend to support studies showing a positive relationship between levels of fear and attitude change but not with respect to behavioral change. Problems dealing with the curvilinear model are discussed.
