Studies have demonstrated that manual and mental rotation show common processes. Training studies have shown that a manual and concurrent visual rotation improves mental rotation performance. In this study, we separated the visual rotation from the manual rotation. In all, 121 participants were randomly assigned to visual training, manual rotation training, or manual training without rotational movement. Before and after the training session of 30 min, they had to solve a chronometric mental rotation test. Data were analysed with linear mixed models and showed an improvement in mental rotation performance for all groups. However, this improvement did not differ between groups. Due to the independence of the form and occurrence of the manual activity, this suggests that it is not the motor activity, but the concurrent visual rotation that leads to improvements in mental rotation tasks. Therefore, the visual component in mental rotation tasks has to be investigated in more detail.