In the last decade, computational models in motor learning have become popular because it provides a theoretical framework not only to explain but predict motor learning behaviors. However, the computational approach sometimes has been criticized by dominating experimentalists in neuroscience due to its lack of the underlying neural mechanisms to support the hypothetical models. We combined computational modeling with a neuroimaging technique such as fMRI to understand human motor memories. First, we provided a unifying mechanism accounting for the spacing and the contextual interference effects using multi-rate motor adaption models. Second, we searched neural substrates related to motor memories predicted by a computational model. We could see the gradual shift of the correlated areas from the fronto-parietal to the anterior cerebellar area with slower time scales. In addition, using multi-voxel pattern analysis (MVPA), we found results supporting separate representation of learning opposing rotations in the cerebellum. Our studies show functional similarity of memory architecture of the brain to that of computer system with fronto-parietal network for a cache memory and anterior cerebellar region for a hard disk.   ​