Visual perceptual learning (VPL) goes through temporal dynamics of consolidation after the offset of visual training. This temporal dynamics of consolidation consists of at least two phases, that is wake-dependent stabilization and sleep-dependent enhancement. After the visual task, VPL becomes stabilized and enhanced during following wakefulness and sleep periods, respectively. This finding is well established based on behavioral results. However, its underlying neural mechanisms have not been clearly identified yet, despite of its importance in understanding the temporal dynamics of VPL. Furthermore, previous findings from the other memory domains including declarative, motor and fear memory suggest that even the previously consolidated memory becomes fragile once again when the memory is reactivated via brief recall. This reactivated memory then becomes reconsolidated with a passage of time. However, it has not been studied yet whether the reactivation and reconsolidation processes also exist in VPL. Studying the reactivation and reconsolidation processes in VPL is important because the finding can reveal how the adult’s visual system changes after training in the long-term periods. In this dissertation study, I investigated three aspects of VPL: (1) underlying mechanism of wake-dependent stabilization, (2) underlying mechanism of sleep-dependent enhancement, and (3) reactivation and reconsolidation processes in VPL. To study these three aspects, I used anatomical and functional magnetic resonance imaging (MRI), magnetic resonance spectroscopy (MRS), magnetoencephalography (MEG), and psychophysics. The results indicate three important things. First, the ratio between excitatory and inhibitory signals (E/I ratio) in the early visual cortex is associated with the wake-dependent stabilization process in VPL. Second, the slow sigma activity corresponding to slow spindles in the early visual cortex is involved in the sleep-dependent enhancement in VPL. Third, reactivation and reconsolidation processes exist in VPL and are associated with the E/I ratio in the early visual cortex. Taken together, the findings reveal how the adult’s visual system balances between stability and instability and changes in the long-term periods after the offset of training.