無論是實驗研究或從現實世異的觀察都發現一般人並非良好的動態決策者，因此如何改善動態複雜問題的決策績效是一個理論與實務上都具重要性的議題。本研究的目的在驗證相同的系統結構下，不同的目標設定可以產生不同的決策行為與績效，進而推論目標設計作為增進動態複雜決策績效的有效工具。本研究設計一電腦化動態複雜決策案例，考量「整體-局部」與「能否表現系統真實狀態」兩個目標構面，比較三種目標設定。研究結果顯示，設定整體考量與能表現系統真實狀能的目標有兩種正面效果使得績效能夠提升，第一是能將決策者的注意力甲導到能反應整體系統狀態的變項上，對提升資訊處理效果有幫助。第二是可以約制偏差決策行為的發生，能減少決策者因無法正確認知決策旁作用之有限理性的負面影響。

　　　　　The studies that improve dynamic decision performance are significant for that people often suffer in dynamic decision environment both in laboratory and field settings. The present study intended to examine that facilitating dynamic decision performance through goal setting was feasible with the premise that goals dictate people's behavior. An experiment using simulation decision task was implemented. Analysis results indicated that the effectiveness of appropriate goal setting has twofold. The first, holistic and representative goals led to better information processing. The second, appropriate goals led subjects to make less dysfunctional decisions derived from people's bounded rationality. The results demonstrated the appropriate design of goals is a viable alternative to improve dynamic decision performance.