To examine whether joint attention enables Japanese monkeys to imitate human actions, we presented an oral action and manual actions directed towards targets to three monkeys who had joint attention through communicative eye-gaze and pointing gestures and to one monkey who had incomplete joint attention and who had acquired imperative pointing but not the use of eye-gaze gestures. Two of the monkeys who were already capable of joint attention were also able to imitate naturally, while the monkey who did not previously show joint attention was not able to imitate until acquiring joint attention capacity. We suggest that joint attention induces natural imitation during interaction between different species--in this case, between monkeys and humans--while individuals not showing joint attention but only attention to a target or movement are only able to follow motion. The monkey may be endowed by nature with motion-following capacity. We speculate that motion-following capacity is developed and controlled through joint attention, and is connected with natural imitation.