This paper describes an approach to the measurement and analysis of Electroencephalogram (EEG) data to develop a brain- computer interface (BCI). The investigation of the serial information processing of human brain is very important for designing brain computer interfaces. Since EEG signals have a good temporal resolution and can be recorded noninvasively from the scalp, EEG has been used as a basic tool of the brain-computer interface. In this study, we measured EEG signals when subjects view, memorize, and recall the simple visual stimuli with directional information. As a result of the frequency analysis for the EEG signals, we found the task-related power change in particular cortical area. Based on this result, we discussed the information processing in the human visual systems and the applicability of the brain computer interface.