User:Jbpassot/Neural computation of voluntary movement

Introduction
We have the ability to control and adapt our movements with a great precision, and to perform efficiently in many different tasks. While this capability does not surprise us on everyday basis, even a simple arm-hand movement to reach for an object is theoretically puzzling. Indeed, we are able to reproduce this simple movement in many different contexts and we master this task without thinking of the constraints we are confronted to : (i) our muscles are sensitive to fatigues and their response from one movement to the next is different; (ii) the neuronal nerves send information at low speed, delaying the sensory feedbacks we could rely on to update and correct our movement; (iii) neural computations often require tens of millisecond, suggesting that most of these feedbacks might be integrated too late to be efficiently used; iv the sensory information is noisy, hence imprecise, and often incomplete (e.g., a task executed in the dark does not have visual feedback); and (v) our body slowly changes through life, and consequently, the properties of our limbs and muscles are not static. All these limitations suggest that our nervous system has been designed to adapt to these constraints and compensate for them.

In this context, the neuroscience of movement relates to the study of the capacity of the human brain to map an abstract task into a set of control laws (i.e., motor commands). Also, the neurocomputational field of movement neuroscience tries to unveil the properties of sensory-motor signals and to highlight the structures responsible for distinct functions related to movement (see for a recent review).