Effects of a therapeutic dose of diazepam (15 mg daily), administered over a 6-week period, were assessed in nonanxious female volunteers. No acute effects of the 5-mg dose of diazepam on performance were observed, but two types of effects were observed on physiology: (1) direct changes on some measures (e.g., skin conductance) and (2) an absence or attenuation of changes expected on the basis of data obtained with a placebo group. The latter effect was found for systolic blood pressure both initially and again after 6 weeks of medication. When performance measured in the morning prior to the first 5-mg dose was compared with that of the morning prior to the last 5-mg dose, 6 weeks later, two types of impairment were found in the diazepam group: (1) decrements in performance (e.g., digit span) and (2) a failure to show expected improvement (e.g., DSST). Performance impairment due to reduced motivation was ruled out by a motivational component of the test battery. Performance tended to return to predrug levels 2 weeks after discontinuation of medication. When physiology was measured in the morning prior to the first dose of the day, 6 weeks of treatment produced decreases in skin conductance, diastolic blood pressure, and heart rate. Blood pressure and heart rate, but not skin conductance, returned to predrug levels 2 weeks after discontinuation of medication. Results of this study indicate that the long-term effects of diazepam can differ from the acute effects and that both acute and long-term effects can take at least two forms, depending upon the behavior or physiological process measured.