A new direction for population vectors

Abstract

There remains a long-standing controversy in physiology about the exact role of primary motor cortex (M1) in the initiation and control of voluntary movement. It is often termed the ‘movement versus muscles’ dichotomy: does M1 control the actual movement of a limb, or the muscle activity that generates those movements? This question has been ‘resolved’ many times, though in different and contradictory ways. Neurophysiological experiments in behaving monkeys have shown that individual M1 neurons are broadly tuned to the direction of a reaching movement; that is, their discharge modulation is greatest for a movement in a particular direction (the cell's ‘preferred direction’, or PD) and decreases only gradually for movements further away from that direction. But despite the coarse tuning of single M1 cells, monkeys can, like us, perform highly accurate limb movements. It has been shown previously that a neuronal population vector – the vectorial sum of directional contributions from a population of M1 cells – correlates closely with the direction of hand movement. These results led to the view that M1 population activity reflects high-level, extrinsic attributes of reaching, namely the trajectory of hand's motion in space, rather than low-level, intrinsic parameters that specify limb mechanics during movement. In one version of the movement versus muscles dichotomy, the former view corresponds to ‘movements’, the latter to ‘muscles’. However, theoretical analyses have demonstrated that, because of the complex coupling between extrinsic and intrinsic movement parameters, the correlation between the population vector and hand trajectory should hold even if M1 cells code muscle activity directly.