Recent in vivo multiunit recording revealed that spontaneous activity (SA) and evoked activity (EA) of neural population in primary visual cortex (V1) show the similar spatiotemporal patterns. Furthermore, the similarity gradually increases depending on visual experience during development. However, it is still unclear how such a network dynamics is organized through the learning of local circuit in cerebral cortex. Through the computational modeling study, it was demonstrated that cortical network could generate precise spike sequences that repetitively occur during both SA and EA, and such the repetitive spatiotemporal patterns are organized through synaptic learning mediated by the spike-timing-dependent plasticity (STDP). As a result, it was found that the model network could acquire the internal representation for external sensory experience described as the spatiotemporal pattern of population firings. It was also provided that the circuit structure for such kind of the visual experience-dependent organization of population dynamics that is recently observed by in vivo multiunit recording study in V1 during development.