Schaal S (2002) Arm and hand movement control. In: Arbib MA (ed) The handbook of brain theory and neural networks, 2nd Edition. MIT Press, Cambridge, MA, pp 110-113 Arm and Hand Movement Control Stefan Schaal sschaal@usc.edu; http://www-clmc.usc.edu Computer Science and Neuroscience & ATR Human Information Systems University of Southern California, 3614 Watt Way–HNB103, Los Angeles, CA 90089-2520 Introduction The control of arm and hand movements in human and nonhuman primates has fascinated researchers in psychology, neuroscience, robotics, and numerous re- lated areas. Movement appears effortless to the uninitiated observer—only when trying to duplicate such skills with artificial systems or when examining the un- derlying neural substrate, one discovers a surprising complexity that, so far, has prevented us from understanding the biological implementation, how to repair neural damage, and how to create human-like robots with a human level of movement skills. Research towards an understanding of motor control can be approached on different levels of abstraction, for instance, by examining the biochemical mecha- nisms of neuronal firing, the representational power of single and populations of neurons, neuroanatomical pathways, the biomechanics of the musculoskeletal system, the computational principles of biological feedback control and learning, or the interaction of perception and action. No matter which level of inquiry is chosen, however, ultimately we need to solve the “reverse engineering” problem of how the properties of each level correlate with the characteristics of skillful behavior. Motor control of arm and hand is an excellent example of the difficul- ties that arise in the reverse engineering problem. Behavioral research has dis- covered a variety of regularities in this movement domain, but it is hard to de- termine on which level they arise. Moreover, most of these regularities were ex- amined in isolated arm or hand movement studies, while, as detailed later, the coordination of arm and hand is a coupled process where hand and arm move- ment influence each other. In this article, we will discuss some of the most prominent regularities of arm and hand control, and examine suggestions where they may come from, with a particular focus on computational and neural net- work models. It will become apparent that an interesting competition exists be- tween explanations sought on a neural, biomechanical, perceptual, and compu- tational level that has created a large amount of controversy in the research community over the years.