The clinical assessment of rigidity is influenced by a number of variables which limit the reproducibility of rating scores and the usefulness of comparisons between subjects. We evaluated an objective measure of rigidity which uses unpredictable but reproducible limb perturbations mimicking the waveform, rate, and amplitude of those used in the clinical examination; and evaluates total resistive force, thus avoiding assumptions about the relative influence of elastic, viscous, or inertial components of the measured resistive forces on the genesis of rigidity. We then used this measure to quantify the effects of an activation procedure on parkinsonian rigidity, because this forms an important but poorly understood part of the routine clinical examination. We studied 20 patients with a clinical diagnosis of Parkinson's disease and 10 age-matched control subjects. A torque motor was used to deliver reproducible, transient, sinusoidal perturbations varying between 1.0 and 1.5 Hz. To quantify rigidity, we calculated angular impulse scores, which reflect the relationship between change in total resistive torque and time. Angular impulse scores were compared with work scores, which have previously been found to correlate with clinical assessments of rigidity. All subjects were studied at rest and with activation. Angular impulse scores were more consistently correlated with rigidity and more clearly differentiated between patients and control subjects than work scores. Activation increased both clinical and objective rigidity scores; activated angular impulse scores ranged from approximately 100%-200% of resting values. When plotted against clinical rigidity scores, activated angular impulse scores lay on a continuum with resting values. We conclude that angular impulse is a valid objective measure of parkinsonian rigidity. Activation increases rigidity, but to varying degrees in different patients. To improve the sensitivity and reproducibility of clinical rigidity assessments, parkinsonian rating scales should include separate resting and activated scores.