Activity of dentate neurons during arm movements triggered by visual, auditory, and somesthetic stimuli in the monkey

1. Single-unit recordings were obtained from 404 neurons in the dentate and interposed nuclei in two monkeys trained to perform simple movements of the elbow in response to three different sensory cues: a light, a tone, and a small, brief perturbation of the trained forearm. Both flexion and extension movements were investigated. 2. Those dentate neurons that showed a clear modulation before the onset of movement (149 of 318 cells recorded) were classified as stimulus related or movement related on the basis of an analysis of the timing of the initial change in discharge. Seventy-one percent of these dentate neurons (106/149) were classified as stimulus related, and 21% (31/149) were classified as movement related. Within the stimulus-related group 87% responded selectively to only one or two of the sensory cues (selective stimulus-related neurons), most often the teleceptive cues, whereas the remaining 13% responded nonselectively to all three cues (nonselective stimulus-related neurons). Interposed neurons, in contrast, showed principally movement-related discharge, and this represented the initial change in discharge in 89% of the neurons. Eleven percent of the interposed cells showed a selective response to the somesthetic cue. 3. The discharge of 28 out of 91 dentate neurons tested with both flexion and extension movements varied with the direction of movement. Few dentate neurons (9%) were found to display any direction sensitivity when considering the discharge preceding the onset of movement, and none of these showed a reciprocal pattern. The discharge of a greater proportion of neurons (24%) was direction sensitive during movement and was occasionally reciprocal. In the same monkeys, however, 78% of the neurons in the neighboring interposed nucleus were direction sensitive, and one-quarter of these displayed reciprocal patterns of discharge. Thus, the discharge of dentate neurons, occurring well in advance of a conditioned movement, cannot specify direction in this simple reaction-time (RT) task. 4. The sensory responses of selective stimulus-related dentate cells ended near the onset of movement but were time locked to the stimulus and not to the movement. When a neuron was responsive to two of the cues the response did not vary with the modality of the stimulus apart from changes in the latencey. The initial sensory response was usually followed by later 'secondary' changes in discharge that were temporally related to the movement. Secondary responses were present in all trials, independent of the modality of the sensory cue and the presence or absence of a preceding sensory response. In a small number of cells, the latency of the sensory response covaried with both the stimulus and the RT. The sensory responses were dependent upon the performance of the conditioned movement, disappearing after movement was extinguished. 5. The modulation of discharge in movement-related dentate neurons was generally later than that in stimulus-related cells. Discharge was also more frequently related to peak velocity than in stimulus-related cells. Direction sensitivity was more frequently observed in the movement-related neurons. 6. The selective stimulus-related dentate neurons showed some localization, being more frequently observed in the caudal two-thirds of the nucleus. Conversely, unmodulated cells were most frequent in the rostral one-third. Only a limited degree of topographical organization in the dentate is suggested by the present study because movement-related and nonselective stimulus-related units were found throughout the nucleus. 7. The discharge characteristics of the selective stimulus-related neurons suggests that this activity may represent a nonspecific triggering mechanism for the initiation of RT movements in response to teleceptive stimuli. The role of these cells is not restricted to the proposed trigger function because most of these cells also showed movement-related activity that was timed so that it could contribute to some aspect of the control of the execution of movement. The movement-related dentate neurons, whose discharge characteristics are closer to the final execution phase than those of the stimulus-related cells, appear to be well suited to a role in controlling the execution of movement.