EFFECTS OF INACTIVATING INDIVIDUAL CEREBELLAR NUCLEI ON THE PERFORMANCE AND RETENTION OF AN OPERANTLY CONDITIONED FORELIMB MOVEMENT

These experiments were designed to examine the effects of inactivating separately each of the major cerebellar nuclear regions in cats on th e execution and retention of a previously learned, operantly condition ed volitional forelimb movement. The experiments test the postulates t hat the cerebellar nuclei, and particularly the interposed nuclei, con tribute substantially to the spatial and temporal features of the inte rjoint coordination required to execute the task and that the engram n ecessary for the retention of this task is not located in any one of t he cerebellar nuclei. All cats were trained to perform a task in which they were required to reach for and grasp a vertical bar at the sound of a tone and move the bar to a reward zone through a template consis ting of two straight grooves in the shape of an invested ''L.'' After the task was learned, the effects of inactivating separately each nucl ear region (the fastigial, interposed, and dentate nuclei) using musci mol microinjections were determined. Data were analyzed by quantifying several features of the movement's kinematics and by determining chan ges in the organization of the reaching component of the movement usin g an application of dimensionality analysis, an analysis that examines the correlation among the changes in joint angles and limb segment po sitions during the task. The retention of the previously learned task also was assessed after each injection. Injections of each nuclear reg ion affected temporal and spatial features of the learned movement. Ho wever, the largest effects resulted from inactivating the interposed n uclei. These effects included an increased length of the reach traject ory, an accentuated deviation of the wrist trajectory from a straight line, cyclic movement of the distal extremity as the target was approa ched, a difficulty in grasping the bar, altered temporal features of t he movement, and a highly characteristic change in the dimensionality measurements. The changes in dimensionality reflected a decreased corr elation (linear interdependence) of the joint angular velocities coupl ed with an increased correlation among the linear velocities of marker s located on the joints themselves. Related but less consistent change s in dimensionality resulted from fastigial injections. The motor sequ ence required to negotiate the template could be executed after the nu clear microinjections, indicating that retention of the motor sequence was not affected by the inactivation of any of the cerebellar nuclei. However, in two of the five animals, some decreases in performance we re observed after dentate injection that were not characteristic of ch anges related to an effect on retention. These data suggest that the c erebellum plays an important role in regulating the consistent, stereo typic organization of complex goal-directed movements, including the t emporal correlation among joint angle velocities. The data also indica te that the retention of the task is not dependent on any of the indiv idual cerebellar nuclear regions. Consequently, these structures are u nlikely to be critical storage sites for the engram established during the learning of this task.