Although the superiority of spaced training over massed training has been established in many forms of learning, the learning efficacy between the two with respect to time efficiency may not be simply compared because a longer total duration of learning is required in spaced training than massed training due to spacing intervals intervening between training sessions in the former. The purpose of the present study was to evaluate the differences in the adaptation of the vestibulo-ocular reflex (VOR) and optokinetic reflex (OKR) after visuo-vestibular training, and to investigate the efficacy of spaced and massed training in mice. Associative visuo-vestibular stimulation was applied to induce VOR and OKR motor learning. Training paradigms were categorized into five groups according to the duration of the spacing interval, keeping the total training time including spacing equal in all training paradigms. Both gain-up VOR training, which increased VOR gain and gain-down VOR training, which decreased VOR gain, increased OKR gain in the massed and spaced learning paradigms. While the increment in OKR gain after gain-up and gain-down training was maintained at 48 h after the end of the last training session, the change in VOR gain by gain-up or gain-down training recovered gradually after training. The OKR adaptation was still in progress during the spacing interval, and the amount of gain increase was greater with longer spacing interval. On the other hand, the VOR gain change after gain-up and gain-down training substantially recovered during the spacing interval. In conclusion, the present study, using learning paradigms with same total duration of training, demonstrated that the spacing effect was more robust in the adaptation of OKR than that of VOR, and the learning effect was maintained longer in OKR than in VOR. These differences in the adaptation of VOR and OKR following identical training conditions suggest that multiple plasticity mechanisms may be differentially involved in the gaze stabilization circuitry.