The sedative effects of a relatively lipid-soluble and a water-soluble beta-blocker were compared in 20 male hypertensives, 30 to 60 years old. In a blinded, randomized, crossover study, critical flash fusion frequency and computerized Stroop Word Test were used to assess psychomotor function parameters during a drug-free control day and then following 14 days of either metoprolol, 150 mg daily, or atenolol, 100 mg daily, treatment. Both drugs caused subtle but significant reductions in both parameters of sedation (critical flash fusion frequency and computerized Stroop Word Testing). Sedation was significantly related to serum concentrations of both drugs. The maximum drug-induced change was 17.2% +/- 9% for metoprolol and 19.6% +/- 3% for atenolol. The duration of effect was six hours after atenolol and two hours after metoprolol. Blood pressure control for all patients was similar during both treatment phases. These results demonstrate that relative lipid solubility does not reliably predict the neurologic effects of beta-blockers. The intensity of drug-induced sedation was similar, but the water-soluble agent produced a longer duration of sedative activity.