In this study we analyzed the verbal and gestural behavior of 6-year-old French children during free dialogue explanation. We expected that children would alter their gestures according to content explained and interlocutor visibility. Thirty children explained two games to a peer: fifteen could not see their interlocutor whereas fifteen could. Results showed that the mean number of clauses per explanation of a child speaker interacting with a child addressee did not significantly change according to content explained or the visibility of the addressee. However, the mean number of gestures per clause is higher for a child speaker interacting with a child addressee both when explaining the spatial game and when face-to-face. Finally, children explaining how to play a spatial game produced more representational gestures and made more interactive gestures when the addressee was visible. These results are discussed in relation to previous studies on children’s language production and acquisition.