背景及目的：雙重作業對平衡的維持是一個相當大的挑戰。過去已有不少研究關於老年人在雙重作業情境下步態變異度的報告，但是對於如何減少平地上的跨步變異度，甚至在不平穩的地面上從事雙重作業的情形則鮮少被探討。本研究的目的乃在探討雙重作業以及增加跨步阻力對不同方向跨步時步伐變異度的影響；此外，還想檢視在不同軟硬度的地面上是否會改變跨步時的步態參數；最後研究者還想探討年齡對不同方向跨步時步伐變異度的影響。方法：本研究選取11位健康的年輕人和11位老年人，老人的迷你智能測試必須高於24分，走路速度必須大於0.8公尺／秒。受試者以正常走路速度分別往前、側、後三方向做跨步動作，此稱為單一作業；雙重作業包括：跨步時同時做算數或逆序背誦12生肖。實驗情況包括在下肢加重或不加重；跨步地面包括一般平地及軟墊。所有測試以隨機的順序進行，研究過程並以數位攝影機與足部開關作資料收集。統計方法使用多變數分析重複型評量。結果：在平均步長表現上，結果顯示老年人跨步步長大於年輕人、雙重作業大於單一作業(p<0.05)。然而在步長變異度的表現上，雙重作業則減少了步長變異度(p<005)。而在不同方向的跨步步長的變異度也有顯著差異(p<0.05)。在擺盪期變異度表現上，老年人在下肢加上阻力之後變異度明顯下降(p<0.05)結論：臨床上在設計物理治療計劃以訓練跨步能力時應可運用雙重作業以增加其挑戰性，而增加跨步阻力則可藉由減少跨步變異度進而增加其穩定度。

Background and Purpose: Dual task paradigm is a challenging situation for balance maintenance. Previous studies have shown that dual task might increase gait variability. However, the effect on stepping ability is not clear. The purpose of this study was trying to investigate whether dual task and adding resistance would affect stepping ability in different direction. Moreover, the effects of age and floor on stepping ability were also investigated. Methods: Eleven normal young and 11 old adults were recruited from Tainan area. All of them were free of neurological diseases. For the older adults, their MMSE scores should be greater than 24 and walking speed greater than 0.8 m/s. For each subject, there were two task situations (single and dual task), two test conditions (with weight and without weight on their dominant leg), two test ground (floor and foam), and three stepping directions (anterior, lateral and posterior) they had to encounter. One digital camera and one foot switch were uses for data collection. Repeated measure ANOVA was use for statistical analysis. Results: Our results indicated that the older adults had longer step length than the young adults. All subjects increased their step length, swing time anti stance time for the dual task trials (p<0.05). However, they decreased their step length variability for the dual task trials. There was a significant difference of step length variability among three stepping directions. Older subjects decreased their swing time variability while they carried weight on their dominant legs (p<0.05). Conclusions: Using dual task paradigm might be a challenging condition for a stepping task. However, adding weight as a proprioceptive stimulation to the dominant leg might decrease the variability of most gait parameters.