Keyboard input for non-alphabetical languages, such as Chinese and Japanese, is problematic because it is labor intensive and imposes a high cognitive load. In our previous work, we measured the effectiveness of handwriting during a note-taking task in Japanese, and found that the input speed during note-taking was higher by hand than by keyboard. The results also showed that the quality of notes taken by hand was higher than that of notes taken by keyboard, and this might have been due to the higher cognitive load during typing. In addition, observation during the experiment revealed several problems subjects faced in the keyboard input task. To evaluate the significance of these observations, we had to obtain quantitative evidence through further study of participant behavior. Therefore, we repeated the experiment, this time with video analysis of the keyboard subtask. By analyzing the participants’ eye movements and their behavior throughout the keyboard subtask we obtained quantitative evidence to support our findings from the previous study. Here, we describe this experiment and our findings in detail.