失眠是最常見的健康問題之一，對患者的生活品質各個層面都會造成相當的影響。失眠的成因複雜而多元，其中心理與行為相關的因素被認為是造成失眠延續為長期問題的主要因素。因此，許多不同的心理及行為治療都被用以治療失眠，近年來整合多種認知及行為技術的認知行為治療（cognitive- behavioral therapy, CBT），更成為治療原發性失眠的主要方法之一。然而，雖然失眠的CBT 療效已獲肯定，但是在其多元的治療方式當中，到底對病患產生什麼改變？哪些改變與失眠的改善程度有關？這些問題都仍然有待澄清。本研究擬收90 名原發性失眠病患，對其進行CBT 團體治療，在治療前後進行預期改變的向度的測量，以這些變項的改變程度作為預測變項，用以預測失眠的療效變項。預計測量的預測變項包含睡前激發程度的生理測量（Heart Rate Variability 與高頻EEG）、激發程度主觀測量（睡前激發程度量表以及貝克焦慮量表）、睡眠的態度與信念（睡眠態度信念量表）以及睡眠相關日常生活行為；療效的效標變項則包含主觀（睡眠日誌與失眠嚴重度量表）及客觀的睡眠改變（polysomnograph 測量）。所得結果將以逐步回歸分別針對不同的療效變項進行統計分析。預計研究結果能增進對於 CBT 改變的變項與失眠療效的關聯性，澄清失眠CBT 的作用機轉，有助於更精確的運用CBT。 Insomnia is one of the most common health complaints in general public; it affects every aspects of the quality of life in affected individuals. The pathogeneses of insomnia is multi-facet. It is now recognized that psychological and behavioral factors play a major role in the perpetuation of insomnia. Thus, many techniques for psychological and behavioral therapies were utilized for the treatment of insomnia. Empirical studies have consistently support the effectiveness of these techniques. Now a day, cognitive-behavioral therapy (CBT) for insomnia that combined several cognitive and behavioral techniques has become one of the major treatments for primary insomnia. Although the effectiveness of CBT for the treatment of insomnia has been confirmed by empirical studies, it is not clear what changes generated by the CBT have contributed to the improvement of insomnia. In the proposed study, 90 patients with primary insomnia will be recruited for 6 weekly sessions of CBT group. The changes targeted by CBT, including presleep physiological status (heart rate variability and high-frequency EEG), subjective rating of arousal level (measured by Pre-sleep Arousal Scale and Beck Anxiety Inventory), sleep beliefs and attitudes (measured by Beliefs and Attitudes of Sleep Scale), sleep-related daily life practices, will be measured prior to the starting and after the completion of CBT group. Changes from pre-test to post-test in these measures will be calculated to serve as predictive variables for treatment outcomes. Treatment outcomes will be measured subjectively by sleep log and Insomnia Severity Inventory, and objectively by polysomnographic sleep recording. Multiple regressions will be conducted to evaluate the predictive power of the different predictive variables for different measures of treatment efficacy. It is expected the findings will help to clarify the mechanisms of CBT for insomnia and will facilitate more efficient ways in the conduction of CBT for the treatment of insomnia.