This paper presents a mathematical model for the learning of accurate
human arm movements. Its main features are that the movement is the su
perposition of smooth submovements, the intrinsic deviation of arm mov
ements is considered, visual and kinesthetic feedback are integrated i
n the motion control, and the movement duration and accuracy are optim
ized with practice. This model is consistent with the jerky arm moveme
nts of infants, and may explain how the adult motion behavior emerges
from the infant behavior. Comparison with measurements of adult moveme
nts shows that the kinematics of accurate movements are well predicted
by the model.