Reaching movements in childhood dystonia contain signal-dependent noise

When reaching, children with dystonia exhibit movements that are slower and more variable than normal children. We hypothesize that in dystonia there is an increase in signal-dependent noise so that there is increased variability with increasing speed. This hypothesis predicts that slower movement in children with dystonia is at least partly due to a compensatory strategy to reduce variability by decreasing speed. To test this hypothesis, we measured the speed of arm movement while children attempted to contact buttons of different sizes. We tested 23 control children and 15 children between the ages of 4 and 16 years with dystonia owing to either cerebral palsy, idiopathic dystonia not due to the DYT1 (torsin A) mutation, or other identified causes. A consistent inverse relationship between movement time and button size was seen for both the control children and the children with dystonia. The variance of movement speed increased with the average speed for all subjects. Children with dystonia moved significantly more slowly at all button sizes, and their movement speed was more sensitive to changes in button size. Therefore, part of the reduction in speed in dystonia is due to relatively greater difficulty in contacting small targets. This finding is consistent with the hypothesis of increased signal-dependent noise in children with dystonia, and we present a simple computational model that provides a possible explanation for the origin of this noise.