本研究主要目的在於探討在不同情緒經驗及反芻風格對男女性自傷行為的影響。回顧過去研究，情緒被認為是多數自傷者自傷的動機及原因之一。可惜的是，大多數研究聚焦在負向情緒經驗，而較少探討正向情緒及不同激發程度的影響力。Selby（2008）的情緒瀑布理論提出反芻風格能夠增加負面情緒強度的論點，也暗示了反芻風格也是另一個與自傷行為相關認知因子。因此本研究認為情緒經驗與反芻風格可能皆為自傷發生的關鍵。有鑑於許多文獻結果指出，男女性的自傷危險因子可能存在差異，本研究便把性別差異納入考量，欲了解男、女性是否在情緒經驗及反芻風格的自傷預測因子有所不同。 本研究對象為508位北部某私立高職部學生，請學生填寫包含基本資料、反芻風格、自傷狀況及情緒經驗的評估共四部份問卷。研究結果顯示，在預測過去一年自傷行為的因子上，的確有性別差異。對女性來說，預測過去一年是否自傷的有效因子分別為負向情緒及負向低激發情緒，男性則是負向情緒及正向低激發情緒。在自傷頻率方面，對男女性來說，沒有任何情緒經驗或反芻風格為有效預測因子。在自傷種類數方面，沒有任何情緒經驗或反芻風格為女性自傷頻率的有效預測因子；對男性來說，則是負向情緒及低激發情緒能夠有效預測過去一年自傷種類多寡。研究最後針對這樣的結果做進一步討論及提供實務上的意涵及應用。 / The purpose of the present study was to investigate the influence of emotion experiences and rumination styles on Non-suicidal self-injury behavior (NSSI) in males and females. According to past studies, emotion was considered a critical factor in motivating individuals to practice NSSI. However, most studies focused on negative emotion experiences, and ignored the role of positive emotions and the effect of emotion activation level. Moreover, Selby’s Emotional Cascade Model proposes that rumination style will increase the negative affect intensity, suggesting that rumination style may be another key factor to NSSI. In addition, due to several studies have shown gender differences in risk factors for NSSI, the gender difference will be also take into account in the present study. To sum up, the present study aimed to examine gender differences in predictors (i.e., emotion experiences and rumination styles) for NSSI. The participants were 508 high school students from one of the private vocational high school in Northern Taiwan. All participants were asked to fill out the questionnaire pack which included demographic questionnaire, short-form rumination style questionnaire, deliberate self- harm inventory, and affect valuation inventory. The results show that gender differences in predicting NSSI existed: First, in predicting the occurrence of NSSI in the past year, negative affect and low arousal negative emotion were the most effective factors for females; negative emotion and low arousal positive emotion were the most effective ones for males. Second, neither emotion experience nor rumination style can predict the frequency of NSSI in the past year for males and females. Third, in predicting the numbers of types of self-injury used in the past year, there was no effective predictor for females; negative emotion and low arousal emotion were the effective predictors for males. Finally, the results and the practical implication were discussed in the thesis.