本研究的主要目的在於以大學生為研究對象，採前瞻式與實驗法兩研究策略來驗 證 Abramson 等人的憂鬱性歸因理論。研究一以貝氏憂鬱量表、日常生活事件歸因量表為研 究工具， 測量 410 位大學生在期中考前後憂鬱情緒的變化與歸因方式，結果發現：由歸因 向度上來探討 Abramson 等人的憂鬱性歸因理論時，在正向事件方面為較不一致的結果，僅 負向事件方面可以得到較一致的相關結果，即個體在面對負向事件時，內在性及概括性兩歸 因向度是非常重要的。以歸因組型的概念作探討時，經驗到正向事件的受試者之歸因組型與 憂鬱情緒無閞，但對於經驗到負向事件的受試者而言，持內在且概括之歸因組型者的憂鬱程 度較持非內在且概括歸因組型者高，此結果部份支持 Abramson 等人的理論。研究二則是針 對研究一的結果，以實驗法作進一步的探究。 研究中以 210 位臺灣大學學生為研究對象， 以貝氏憂鬱量表、Peterson 等人所編製的歸因量表為研究工具， 操弄個體在概念形成認知 作業上所做行為與所得結果間的關聯性，針對習得無助感的兩個指標（憂鬱程度及認知能力 ）進行 2 × 2 × 2 的三因子變異分析，此三因子為（ 1 ）歸因組型（分為內在概括組及 非內在概括組），（ 2 ）關聯組別（分為關聯組與非關聯組）及（ 3 ）測量時間（實驗前 與實驗後各測量一次）；前二變項為受試者間變項，後一變項為受試者內變項。所得研究結 果如下：個體會因歸因組型的不同而有不同的憂鬱情緒變化，且個體會因對事件產生不同的 不可控制知覺而有不同的情緒變化；受試者在答題反應時間上會因客觀關聯性的不同而有不 同。此結果支持 Abramson 等人理論中對於關聯性的預測。本文亦就研究結果顯示的意義與 未來可能的研究方向提出討論。

　　　　　In line with Abramson et al's reformulated learned helplessness theory and hopelessness theory of depression (Abramson et al, 1978; 1989), the purpose of this study was to test diathesis-stress models by both prospective and experimental design in college students. In study Ⅰ, we tested the prediction with a prospective designs by determining whether the depressive attributional styles before the midterm exam predicted the severity of their depressive mood response after receiving their grade on a midterm exam. The results partially supported the reformulated theory: When students faced positive outcome, their mood was not consistent with the prediction; whereas when negative outcome was encountered, the change of mood was as predicted specifically, the results indicated that the internal-external attributional dimension and the global-specific attributional dimension were crucial to depressive mood. According to the findings of Study Ⅰ, Study Ⅱ further tested the hypothesis using an experimental design by manipulating the contingency between students' behavior and their outcome in order to understand the relationship between their attributional style and their mood. The results revealed that when facing a negative event, students with an internal-and-global attributional style would experience depressive mood, which was consistent with the theory, and the "uncontrollability" was also an important factor. The implications of the present findings and future research possibilities are postulated.