Selection and coordination of human locomotor forms following cerebellar damage

We have previously shown that control subjects use two distinct temporal st rategies when stepping on an inclined surface during walking: one for level and 10 degrees surfaces and another for 20 and 30 degrees surfaces. These two temporal strategies were characterized by systematic shifts in the timi ng of muscle activity and peak joint angles. We examined whether cerebellar subjects with mild to moderate gait ataxia were impaired in their ability to select these two temporal strategies, adjust peak joint angle amplitudes , and/or adjust one joint appropriately with respect to movements and const raints at another joint. Subjects walked on a level surface and on differen t wedges (10, 20, and 30 degrees) presented in the context of level walking . In a single trial, a subject walked on a level surface in approach to a w edge, took a single step on the wedge, and continued walking on an elevated level surface beyond the wedge. Cerebellar subjects used two temporal stra tegies, one for the level and 10 degrees surfaces and another for 20 and 30 degrees surfaces. Cerebellar strategies were similar to those used by cont rols except for the timing of ankle-joint movement on the steeper wedges. C erebellar subjects adjusted the peak amplitudes of individual joint angles normally, with the exception of peak ankle plantarflexion. However, they ex hibited greater trial-to-trial variability of peak hip and knee joint angle s that increased as a function of wedge inclination. The most substantial d eficit noted in the cerebellar group was in the relative movement of multip le joints. Cerebellar subjects demonstrated multijoint coordination deficit s in all conditions, although these deficits were most pronounced during st ance on the steeper wedges. On the 30 degrees wedge, cerebellar subjects sh owed abnormal relative movement of hip, knee, and ankle joints and the most substantial decomposition of movement. We speculate that to simplify multi joint control, cerebellar subjects decomposed their movement by fixing the ankle joint in a dorsiflexed position on the steepest wedges. Our results s uggest that the cerebellum may not be critical in selecting the basic motor patterns for the two temporal strategies because cerebellar subjects produ ced appropriate timing shifts at most joints. Instead, our data suggest tha t the cerebellum is most critical for adjusting the relative movement of mu ltiple joints, especially to accommodate external constraints.