PRIMATE BASAL GANGLIA ACTIVITY IN A PRECUED REACHING TASK - PREPARATION FOR MOVEMENT

Single cell activity was recorded from the primate putamen, caudate nu cleus, and globus pallidus during a precued reaching movement task. Tw o monkeys were trained to touch one of several target knobs mounted in front of them after an LED was lighted on the correct target. A precu e was presented prior to this target ''go cue'' by a randomly varied d elay interval, giving the animals partial or complete advance informat ion about the target for the movement task. The purpose of this design was to examine neuronal activity in the major structures of the basal ganglia during the preparation phase of limb movements when varying a mounts of advance information were provided to the animals. The reacti on times were shortest with complete precues, intermediate with partia l precues, and longest with precues containing no information, demonst rating that the animals used precue information to prepare partly or c ompletely for the reaching movement before the target go cue was given . Changes in activity were seen in the basal ganglia during the prepar atory period in 30% of neurons in putamen, 31 % in caudate nucleus, an d 27% in globus pallidus. Preparatory changes were stronger and more c losely linked to the time of movement initiation in putamen than in ca udate nucleus. Although the amount of information contained in the pre cues had no significant effect on preparatory activity preceding the t arget go cue, a directional selectivity during this period was observe d for a subset of neurons with preparatory changes (15% in putamen, 11 % in caudate nucleus, 14% in globus pallidus) when the precue containe d information about the upcoming direction of movement. A smaller subs et of neurons showed selectivity for the preparation of movement ampli tude. A larger number of preparatory changes showed selectivity for th e direction or amplitude of movement following the target go cue than in the delay period before the cue. The intensity of preparatory chang es in activity in many cases depended on the length of the delay inter val preceding the target go cue. Even following the target go cue, the intensity of the preparatory changes in activity continued to be sign ificantly influenced by the length of the preceding delay interval for 11% of changes in putamen, 8% in caudate nucleus, and 18% in globus p allidus. This finding suggests that preparatory activity in the basal ganglia takes part in a process termed motor readiness. Behaviorally, this process was seen as a shortening of reaction time regardless of p recue information for trials in which the delay interval was long and the animals showed an increased readiness to move. Preparatory activit y in putamen following the target go cue was most intense in trials wi th a short delay interval, in which motor readiness had not achieved i ts maximum level prior to the go cue. The results of this study indica te that the basal ganglia are involved in multiple aspects of preparat ory processing for limb movement. Preparatory processing is therefore unlikely to be divided anatomically along the functional lines examine d in this study. In the basal ganglia, preparatory processing reflects both preparation for target selection and control of timing the onset of movement (motor readiness). These characteristics can be integrate d in a functional scheme in which the basal ganglia are predominantly responsible for the automated execution of well-trained behavior.