Neural control of limb movement
Many chronic unit recording studies have documented how neurons covary with movement parameters, although simple interpretations of neural function in terms of explicit representation are problematic [1]. Moreover, such covariations cannot establish causal relationships. To identify those cells that have causal effects on muscles we used spike-triggered averages of muscle activity to document post-spike effects of cells on motoneurons (Fig. 1). By knowing both the response patterns of these cells during movements and their output connections to target muscles we can make important causal inferences about their contribution to movements. Premotor cells were documented in motor cortex [2], red nucleus [3], dorsal root afferent fibers [4] and spinal interneurons [5]; all groups showed characteristic response patterns during alternating ramp-and-hold wrist movements [6] (Fig. 2). Motor cortex cells with correlational linkages to muscles (CM cells) affected multiple target muscles and had specific relations to movements; in contrast, spinal cord interneurons had smaller muscle fields and were active through broader ranges of movement (Fig. 3).
References