It is thought that we can discriminate between animate and inanimate things. This ability is called animacy perception. Our discrimination between animate and inanimate things is considered to be an important ability for our social cognition, because animacy perception is assumed to serve as a foundation for considering objects as others that have their own goals, intentions and⁄or emotions. We investigated neural mechanism underlying animacy perception using a real animate thing (turtle) and an inanimate thing (robot) in this study. As far as we know, brain activity related to animacy perception in the course of approaching a real animate thing has not been investigated. In experiment 1, we compared Event-Related Potentials (ERPs) when participants performed reaching actions for the animate thing (animate condition) with those when they performed reaching actions for the inanimate thing (inanimate condition). We found that the amplitude of ERPs in left infero-frontal region, which is said to correspond to mirror system, was significantly higher in the animate condition than in the inanimate condition. Moreover, we found more significant mu suppression in the animate condition than in the inanimate condition, which is said to be an evidence of the activation of mirror system. These results suggest that mirror system is related to animacy perception. In Experiment 2, participants were asked to observe an object, either of the animate thing or the inanimate thing, which was covered in a box so that they could not judge by appearance what it was and to answer whether they felt it as animate or inanimate. We compared ERPs when they performed reaching actions for the objects that they felt as animate with those when they performed reaching actions for the objects that they felt as inanimate. As well as in Experiment 1, we found that the amplitude of ERPs in left infero-frontal region was significantly higher in the former condition than in the latter one. In conjunction with the results in Experiment 1, this result suggests that the activation of mirror system takes a role in subjective attribution of animacy to objects.