本研究主要目的在於操弄回饋內容及特定範圍回饋安排，以檢驗學童對時間性動作技能的學習效應。研究以60位12.3 ± 0.3歲兒童為實驗參加者，隨機分配至二種回饋內容（整體時間、分段時間）及三種回饋安排（BW0%、BW5%、BW10%）等6個組別，進行N字形按鍵工作的練習，並要求三段動作的時間分別為200毫秒、600毫秒與400毫秒，總時間為1200毫秒。通過變異誤差、恆常誤差、絕對誤差及相對時宜誤差等變數計算後，以2 × 3 × 8與2 × 3 × 3混合設計三因子變異數分析進行統計考驗，並以Duncan法進行事後比較，顯著水準為.05。結果顯示，各組參加者在穩定性與準確性的動作表現，均隨練習量增加而改善，且特定範圍回饋安排在獲得期的恆常誤差低於每次給回饋的組別（p < .05）。5%與10%特定範圍回饋安排在保留測驗的絕對誤差也均低於每次給回饋的組別（p < .05），而分段時間回饋組別的恆常誤差顯著低於全部時間回饋組（p < .05），顯示10%特定範圍回饋安排與分段時間回饋均為有利於兒童按鍵工作的學習變項。

　　The purpose of this study was to investigate the learning effects of timing motor skill through the manipulations of content of knowledge of result (KR) and bandwidth KR in school children. Sixty participants, 12.3 ± 0.3 years old, were assigned randomly to one of groups of two KR contents (overall or segments movement time) and three bandwidth KR schedules (0, 5, or 10% bandwidth condition) to practice the pressing task of N shape in total time of 1,200ms with the segment duration of 200ms, 600ms and 400ms respectively. The variable error (VE), constant error (CE), absolute error (AE), and relative timing error (AEprop) were calculated and adopted to analyze the statistical differences in 2 × 3 × 8 and 2 × 3 × 3 mixed-design three-way ANOVAs and Duncan post-hoc test with an alpha level of .05. The results showed that six groups had the tendency to improve accuracy and stability during acquisition. The CE showed 10% BW was less significantly than 0% BW during acquisition (p < .05). During retention test, AE in 5% and 10% BW conditions were lower than 0% BW, and CE in the condition of segments movement time feedback was lower than overall (p < .05). This study indicated that 10% BW scheduling and segments movement time feedback were the learning variables to benefit children to learn the pressing task.