運動行為的產生，是知覺與運動互動的結果，學習運動技能的過程中，根據運動性質、情境的不同，不同知覺管道對運動學習結果的影響也有不同。目的:比較視覺與聽覺兩種知覺管道對於學習新運動技能的影響。方法:採用虛擬實境方式，利用電腦模擬類似真實運動情境，以十六名大學生為實驗參與者，分為視覺組、聽覺組各八人，分別使用頭戴式監視器或耳機來操作向上擊球的動作。結果:顯示視覺組與聽覺組的練習表現達顯著差異F(1，14)=5.977 ' p< .05 '練習效果亦達顯著差異F(2， 28)=12.527 ' p< .05 ，知覺與練習區段交互作用有顯著差異F(2， 28)=5.747 ' p< .05。進行單純主要效果比 較發現:(一)前兩天視覺與聽覺未達顯著差異;(二)最後兩天視覺與聽覺達顯著差異;(三)遷移測驗視覺與聽覺達顯著差異;(四)視覺組與聽覺組之前兩天、最後兩天、遷移測驗皆達顯著差異，再經事後比較結果顯示:視覺組前兩天與最後兩天、遷移達顯著差異;聽覺組前兩天與遷移達顯著差異。結論:單一視覺知覺在虛擬實境情況中，學習向.上擊球比聽覺知覺有較佳的幫助，而在虛擬實境的運動學特徵，較佳的擊球動作表現，其擊球點之加速度為負值，與實際擊球運動學特徵類似。

Motor learning is a process associated with practice or experience leading to relatively permanent changes in the capability for movement (Schmidt & Lee， 1999). Perception and action are mutually influenced. Purposes: To compare the effects of visual perception and auditory perception on motor learning and to examine the kinematic characteristics of bouncing a ball in virtual condition. Methods: Using computer simulation to interact with human movement to study the effect of vision and audition on learning. Sixteen undergraduate students were assigned to either vision or audition group. Participants practiced 5 trials of the virtual hitting task a day for 8 days followed by 2 days' transfer test where both visual and auditory information were provided. A 2 (vision and audition group) x 3 (the first two days， the 7th and the 8th days， and the 9th and 10th days) mixed design ANOVA on the number of hit was used to analyze the effects of the perceptual modes and learning. Results: Effects of perceptual mode， practice block and the interaction of the two are all significant (p< .05). The simple main effect shows that there is a significant difference between visual and auditory groups. Post hoc analyses show that for the vision group， the first block is significantly different from the second and the transfer block; for the auditory group， only the first block is significantly different from the transfer test. Conclusion: Visual information appears to be more useful than auditory information in learning the virtual hitting task. The kinematics data indicates that the performance is more stable when the bat shows negative acceleration at the impact point which is similar to hitting the real ball in the natural environment.