In order to investigate the contribution of the vestibular system to spatial orientation, we studied memory-guided saccades in three conditions: visual-memory guided saccades (ViC), saccades to the remembered spatiotopic position of a visual target, after whole-body rotation (SVeC) and saccades to the remembered retinotopic position of a visual target, after whole-body rotation (RVeC). Visual feedback presented after each trial allowed eye position correction. The error was larger in SVeC, but the performance improved throughout the experiment (learning) in that condition only. As learning occurred over the first four trials, we omitted these trials from the average computation, and the significant difference between the conditions disappeared. It is concluded that vestibular information does contribute to update the internal spatial representation of visual information when a visual feedback is provided.