The goals of this study were to determine whether functional outcome after motor training in rats was linked to synaptic plasticity in thalamus, and whether the Rota-rod apparatus, widely used to test motor function, could be used as an easy and quantitative motor skill training procedure. Adult female Sprague-Dawley rats (n = 39) were evaluated under three training conditions: 1. Movement requiring balance and coordination skills on Rota-rod; 2. simple exercise on treadmill; 3. nontrained controls. Motor function was evaluated by a series of motor tests (foot fault placing, parallel bar crossing, rope and ladder climbing) before and 14 or 28 days after training procedure. Synaptic strength in brain was assessed by synaptophysin immunocytochemistry. After 14 days of training, Rota-rod-trained animals significantly (p < 0.01) improved motor performance, compared to treadmill and nontrained animals. Animals with up to 28 days of simple exercises on the treadmill did not show a significantly improved performance on most motor tasks, except for an improvement in foot fault placing. Intensive synaptophysin immunoreactivity was present in the right but not the left mediodorsal and ventromedial nuclei of thalamus in Rota-rod-trained rats at 14 and 28 days, and in treadmill-trained rats at 28 days. The data suggested that functional outcome is effectively improved by motor skill training rather than by simple exercises, and this may be related, at least partially, to uniquely lateralized synaptogenesis in the thalamus. Both Rota-rod and treadmill could be quantitatively used in rats for motor training of different complexity.