exercise has consistently been found to reduce anxiety (petruzello, landers, hatfield, kubitz, & salazar, 1991). however, the mechanisms underlying the effect of exercise on anxiety are unknown. the present study evaluated the expectancy hypothesis, which posits that the anxiolytic effect of exercise is due to the expectation of such benefits from participating in exercise. it was hypothesized that expectancy would have no effect on the anxiolytic effects of exercise. ninety participants (56 females and 34 males) were randomly assigned to one of three groups; 1) an exercise condition (n=31), 2) a quiet rest condition (n=31), and 3) a no expectancy exercise condition (n =28). state anxiety was measured at baseline, after anxiety manipulation, and following the assigned group condition. participants in the two exercise conditions experienced a slight decrease in state anxiety while participants in the quiet rest control group experienced a slight increase in state anxiety. although these changes in state anxiety did not reach significance, participants in the two exercise groups experienced an overall decrease in anxiety which was significantly different from the quiet rest group which experienced an increase in anxiety. these findings support the hypothesis that the anxiety reducing benefit of exercise is unlikely to be due to expectations.