Temporal coordination of behaviors is observed in various social communications. Kawasaki et al. (2013) reported that EEG alpha exhibits inter-brain synchronization more with better temporal coordination in a two-person alternate speech task. The mechanism why EEG inter-brain synchronization emerges in communication tasks is an open question. In this paper, we aim to elucidate the temporal coordination by using a phase-oscillator model with phase response curve (PRC). The pair of subjects was instructed to make their inter-tapping-interval (ITI) as similar as possible to the preceding ITI made by each partner's tapping. A signal with subject-specific color was shown to both the subjects after each tap. This task was conducted in human-human (HH) and human-computer (HC) conditions. The performance measure in term of the matching rate was defined as the ratio of tap trials where the difference between subject' ITI and the preceding one is less than a value. In HC condition, each subject played with a computer program implemented to reply its ITI same as the preceding one plus the Gaussian noise. In HC condition, the subject tends to keep almost the same ITI, i.e., the same oscillation phase and same period. The oscillation phase shift forced by the noise enabled to measure phase response curve over all values of the oscillation phase. 19 pairs (38 subjects) participated in the experiment. PRCs were found either the types 0 or 1, suggesting different strategies in temporal coordination in different subjects. In addition, the questionnaire of a psychological test was applied to the subjects. The score on the communication ability in he questionnaire showed correlation with the matching rate in HC condition. Thus, our findings imply the importance of temporal coordination in social communication(p<0.05). The link between behavioral coordination and neural dynamics related with social communication should be elucidated at the next step.