AbstractCortical activation is reduced when patients with focal dystonia perform movements that do not induce dystonic posturing. This finding suggests that the cortical drive to muscles may in some circumstances actually be reduced not increased, as suggested by basal ganglia models of dystonia as a hyperkinetic disorder. The purpose of this study was to examine flexor and extensor strength at the wrist (a clinically affected joint) and elbow (a nonclinically affected joint) in 18 patients with primary focal hand dystonia compared to matched control subjects. We measured peak torque from maximum voluntary contractions, and agonist and antagonist muscle activation by means of surface electromyograms. Patients were significantly weaker than controls at both the elbow and wrist joints and in both flexors and extensors compared to controls. Peak elbow flexion torque was, on average, 14.4% lower in the dystonic compared to the control group, elbow extensor peak torque was 28.6% lower, wrist flexor peak torque was 17.4% lower, and wrist extensor peak torque was 20.7% lower. Strength did not differ as a function of clinical severity. Reductions in peak torque were accompanied by reduced agonist activation, although this finding only reached statistical significance at the elbow. The amount of co‐contraction of antagonistic muscles was not significantly different between the two groups. These results are discussed in the context of dystonia as a disorder resulting from dysfunction of basal ganglia output. © 2005 Movement Disorder Society