Misattribution, preparatory information, and speech anxiety.

In two experiments with undergraduate subjects, I compared the effects of misattribution versus information manipulations on speech anxiety. In Experiment 1, some subjects were allegedly exposed to subliminal noise while reading a speech in front of a camera. These subjects were told that subliminal noise makes people feel either unpleasantly aroused or pleasantly relaxed or that it has no effect. Subjects in a fourth condition were given accurate information about how they would feel (unpleasantly aroused) but were not exposed to the subliminal noise misattribution source. In Experiment 2, I replicated the arousing noise and accurate information conditions from the first study and added two new groups incorporating a delay that should preclude misattribution. In both experiments, the alleged presence of arousing subliminal noise reduced subjects' speech dysfluencies during the speech task, whereas the presentation of accurate information alone did not have a comparable ameliorative effect. Thus, both experiments supported the misattribution interpretation of why neutral labels for arousal can reduce emotionality.