Theories of motor control postulate that the brain uses internal models of
the body to control movements accurately. Internal models are neural repres
entations of how, for instance, the arm would respond to a neural command,
given its current position and velocity(1-6), Previous studies have shown t
hat the cerebellar cortex can acquire internal models through motor learnin
g(7-11). Because the human cerebellum is involved in higher cognitive funct
ion(12-15) as well as in motor control, we propose a coherent computational
theory in which the phylogenetically newer part of the cerebellum similarl
y acquires internal models of objects in the external world. While human su
bjects learned to use a new tool (a computer mouse with a novel rotational
transformation), cerebellar activity was measured by functional magnetic re
sonance imaging, As predicted by our theory, two types of activity were obs
erved. One was spread over wide areas of the cerebellum and was precisely p
roportional to the error signal that guides the acquisition of internal mod
els during learning. The other was confined to the area near the posterior
superior fissure and remained even after learning, when the error levels ha
d been equalized, thus probably reflecting an acquired internal model of th
e new tool.