Tested the notion that dissonance has arousal properties in a 2 * 3 design with 77 male college freshmen. 3 groups of Ss were induced to write counterattitudinal essays under either high- or low-choice conditions. Group 1 was led to believe that a pill, which they had just taken in the context of a separate experiment, would lead them to feel tense. Group 2 was led to believe that the pill would cause them to feel relaxed. Group 3 expected the pill to have no side effects whatsoever. In this last condition, the results yielded the usual dissonance effect: high choice produced more attitude change in the direction of the essay than low choice. When Ss could attribute their arousal to the pill, this effect was virtually eliminated. When Ss felt they should have been relaxed by the pill, this effect was significantly enhanced. Implications of these results for L. Festinger’s 1957 statement that dissonance is a drivelike state are discussed.