歷年博碩士論文

大專男性追求肌肉發達的身體意象與自尊、生活品質之探索研究

Exploring Trend in Male Body Image and Relationships among Drive for Muscularity, Self-Esteem, and Quality of life

摘要

近代，西方學者觀察到各式媒體傳遞出男性在意身體外貌的新趨勢，相對於鼓勵女性追求纖瘦體態，而是強調男性應擁有肌肉發達的體態。現階段，已有西方的研究指出男性過度追求肌肉發達的體態與其身心功能呈現負向的關聯。可見，在社會變遷下，身體意象也漸在男性心理適應上扮演重要角色。但，國內關於男性身體意象的相關研究仍屬闕如、可茲使用的測量工具亦顯不足，且社會文化是否傳遞出男性在意體態的背景資料也較有限。因此，本研究藉探尋男性雜誌中男體裸露的變化趨勢，作為男性可能受到社會影響漸漸在意體態的背景資料；然後發展適切之中文版工具與研究程序，探索男性在意身體外貌的趨勢，會如何反應在追求肌肉發達體態，與自尊、生活品質等心理社會功能的關聯上，最後檢驗自尊與男性追求肌肉發達的身體意象對於其生活品質的預測力和路徑模式。

研究一以具一定閱讀率之男性雜誌，抽樣統計流行風尚單元在八年間男體裸露程度的變化；研究二以190名大學男性為樣本，除檢視本研究所編製中文版追求肌肉發達量表與中文版多向度身體自我關係問卷五個分量尺之心理特性之外，也檢驗男性追求肌肉發達的身體意象與自尊、生活品質的關聯性，並以路徑分析探索其互動模式。

結果顯示：（一）近十年內，隨著時間演進，雜誌中男性身體變得更加裸露。（二）中文版追求肌肉發達量表與中文版多向度身體自我關係問卷五各分量尺皆具良好之心理計量特性。（三）外貌、體適能與身體滿意等身體意象皆與男性自尊與生活品質呈顯著正向關聯。（四）男性追求肌肉發達之情感評價與自尊呈顯著負向關聯，而行為取向則與生活品質呈顯著正向關聯。（五）路徑分析顯示，自尊可直接影響身體意象對生活品質的影響度，也可透過追求肌肉發達的身體意象部份中介而影響之。

最後以所建構之模式探討增進男性身體意象與可能之臨床應用，同時也報告未來納入男性氣概與媒體影響的可行研究方向。

Abstract

Recently, western researchers observed a growing trend of putting more emphases on the appearance of male body in a variety of media. Specifically, men believe that they should have muscular bodies rather than focus on drive for thinness as women do. Studies also indicate that there is a negative association between over concern for muscularity and psychological functions in western men. As the societies change, body image becomes more and more important for male’s psychological well-beings. However, rare studies related to male body image have been done in Taiwan, and the measurement tools are limited either. Hence, the aims of this study were to first explore the trend of male body nudity in a popular magazine in Taiwan as the background information for this study; secondly, to develop tools to measure male body image; and finally, to investigate the relationship between men’s drive for muscularity and self-esteem, as well as their predictions for quality of life.

The present thesis consists of two studies. In Study 1, a rating system of six-degree body nudity was developed and then applied to evaluate the male figures appeared in the fashion section of a leading men’s magazine from 1998 to 2005. In Study 2, psychometric properties of five subscales of the Chinese version of the Multidimensional Body-Self Relation Questionnaire (CMBSRQ) and the Chinese version of the Drive for Muscularity Scale (CDMS) were examined by using samples of 190 male undergraduate students. Furthermore, it not only investigated the relationships among drive for muscularity, self-esteem, and quality of life, but also examined the integrative correlational model by path analysis.

Results revealed that: (1) a significant correlation between the year of publication and the degree of nudity was found. (2) Both the CMBSRQ and CDMS yielded adequate and satisfactory psychometric properties, supporting themselves as useful measures of body image. (3) All the five subscales of CMBSRQ significantly and positively correlated with self-esteem and quality of life. (4) Evaluative component of drive for muscularity related to poor self-esteem, and behavioral component related to positive quality of life. (5) Path analysis showed that self-esteem affected quality of life directly and the process was partially mediated by the two components of drive for muscularity.

Based on the findings, further discussion was about how to apply it to improve male body image or use it in clinical setting. Various factors such as masculinity and media were also suggested for future research.

憂鬱反芻對工作記憶容量的影響之初探

Investigating the Effect of Depressive Rumination on Working Memory Capacity in Non-Clinical Samples

摘要

前言：過去文獻顯示憂鬱反芻是維持個體憂鬱狀態的重要關鍵，然而近來研究指出憂鬱反芻的概念可區分為苦惱自責和深思反省兩個因素，分別對應於反芻思考的不適應性以及適應性的面向。另一方面，以認知功能角度探究反芻持續的可能作用機制，目前有反芻傾向為認知功能缺損者的外顯表徵，以及反芻歷程耗費個體認知資源兩種觀點，但都有待後續研究加以驗證。目的：本研究企圖以實徵研究方式初探反芻傾向、反芻狀態與工作記憶容量之關聯。

方法：研究一旨在發展憂鬱反芻反應的測量工具，以220 名大學生為樣本，檢驗中文反芻反應風格量表短版在本地的心理計量特性與因素結構，並驗證憂鬱反芻兩因素模型。研究二旨在發展中文操作廣度作業，建構難度相當的平行版本作為後續研究的工具，以測量個體在接受操弄前後之工作記憶容量。研究三包含兩個實驗，實驗一以70名大學生樣本檢驗反芻思考是否消耗認知資源。實驗二以另外58 名大學生為受試者，在反應作業操弄前進行悲傷情緒的誘發，初探憂鬱反芻對於認知功能的影響。

結果：綜整研究結果顯示：（1）中文反芻反應風格量表短版具有良好的心理計量特性，並且苦惱自責分量表與憂鬱症狀嚴重度有顯著正相關，但深思反省分量表則與之無關，支持憂鬱反芻為多向度之概念。（2）反芻思考歷程本身無法引發負面情緒或造成受試者認知功能的缺損。（3）一般受試者在誘發悲傷情緒後進行反芻作業，並未展現情緒或執行功能上的負面效果。值得注意的是，深思反省傾向高分組在悲傷情緒復原後，工作記憶容量表現顯著優於低分組。

討論：最後以思考內容的情緒價、樣本特性等結構層面探討憂鬱反芻理論的應用範圍，並說明研究結果在臨床應用上的啟示，以及未來可行的研究方向。

Abstract

Introduction: Literature review shows that depressive rumination sustains depressed mood. However, recent studies have suggested of two separated subcomponents of depressive rumination, i.e., brooding and reflection, representing the maladaptive and adaptive cognitive aspects of depressive rumination, respectively. From another point of view, some researchers proposed that there were cognitive mechanisms accounting for the persistence of rumination process, i.e., the tendency to ruminate is the results of individual’s cognitive impairments or rumination itself occupies one’s cognitive resources. Both hypotheses need to be further examined.

Purpose: Thus, the present thesis aimed to empirically investigate the relationship among rumination tendency, rumination process and working memory capacity. Method：Study 1 examined the psychometric properties of the Chinese version of the Rumination Response Scale-Short Form (CRRS-SF). It validated a two-factor model of depressive rumination by using a sample of 220 undergraduate students. Study 2 developed the Chinese version of operation span task, and resulted in two parallel sets of items equivalent in difficulties for evaluating subjects’ working memory capacity pre- and post- manipulation in the next study. Study 3 consists of two experiments. Experiment 1 was designed to examine whether the rumination process occupied cognitive resources in 70 undergraduate students. In experiment 2, another 58 students were instructed to induce sad mood before manipulation in order to further investigate the impact of depressive rumination on cognitive function.

Results: All in all, the results indicate that: (1) The CRRS-SF yielded adequate psychometric properties. Moreover, scores on the brooding subscale, rather than reflection subscale, were significantly positively correlated with depressive symptoms. The results supported a satisfactory two-factor construct of depressive rumination. (2) Rumination process alone, in the absence of depressed mood, was not able to elicit participants’negative emotions or impair their cognitive function. (3）Normal subjects with sad mood induction did not show deleterious effects on mood or executive function after ruminating. Notably, participants with higher scores on reflection subscale showed greater working memory capacity than participants with lower scores after remediation from sad mood states.

Discussion: Based on the findings, the structure aspects such as the valence of thought contents and subjects’ characteristic were discussed to elucidate the appropriateness of applying the theory of rumination in non-clinical samples. Possible application and future directions were also addressed.

初探兒童身體虐待及創傷後壓力疾患的盛行率和危險因子：以四到八年級台灣學生樣本為例

Exploring Child Physical Abuse and PTSD in a Sample of Taiwanese Fourth to Eighth Graders: Prevalence and Risk Factors

摘要

目的　本研究目的在調查台灣兒童及青少年社區樣本中，兒童青少年身體虐待及其相關之創傷後壓力疾患的盛行率，並檢驗人口變項、受創者的主觀感覺、及創傷事件特徵對創傷後壓力疾患的影響。

方法　參與者年齡分佈為9-15歲。所有人均在班級團體施測情境下填答修改版的UCLA PTSD Reaction Index for DSM-IV (兒童及青少年版)，藉以調查其遭受身體虐待的頻率、發生距今時間、加害者、以及與身體虐待事件相關的創傷後壓力疾患症狀。

結果　分析結果顯示，兒童身體虐待的終身盛行率為34%，男童(38.1%)的比率高於女童(29.8%)，以父母為最常見的加害人。在曾經遭受身體虐待的樣本中，13.6%現階段仍符合創傷後壓力疾患之診斷，另外16.9%符合部分診斷。其中，性別和發展階段的差異均不顯著。危險因子的調查發現，主觀覺知威脅、遭受非父母之成人的身體虐待經驗、身體虐待之頻率、以及診斷準則A2均顯著預測創傷後壓力疾患。

討論　本研究發現台灣樣本的兒童身體虐待頻率遠高於西方族群，但其後的創傷後壓力疾患症狀卻相對較少。華人文化的教養和教育觀、以及人際衝突解決模式的影響將在文中探討。

Abstract

Objective: The study aimed to report the lifetime prevalence of child physical abuse (CPA) and current prevalence of full/partial PTSD following CPA in the community sample of Taiwanese fourth to eighth graders. Risk factors of PTSD, including demographic factors, victims’ subjective reactions, and event-related characteristics, were examined.

Method: All participants, aged 9 to 15 years old, completed the modified version of UCLA PTSD Reaction Index for DSM-IV in class setting. The frequency, time elapsed, and perpetrators of their CPA and the following PTSD symptoms were assessed.

Results: The lifetime prevalence of CPA was 34%. Males exhibited higher prevalence (38.1%) than females (29.8%). Parents were the commonest perpetrators. Of those who had experienced CPA, 13.6% and 16.9% met full and partial PTSD, respectively. No gender and grade difference was found. Perceived threat, experience of CPA by non-parent adults, CPA frequency, and criterion A2 were the major risk factors for PTSD.

Discussion: CPA prevalence in Taiwan was higher than in the West, while posttraumatic symptom prevalence was lower. The influence of Chinese culture, by the ways of parenting and educational styles and strategies for coping with interpersonal conflicts, were discussed.

憂鬱症患者人際認知模式之探討：憂鬱人格傾向與臉部情緒認知基模反應

A Clinical Investigation of Cognitive Interpersonal Model of Depression: Personality Predispositions and Schematic Responses to Facial Emotions

摘要

憂鬱的認知理論認為憂鬱衍生基模經由壓力事件的活化造成憂鬱的初發及維持，人際理論則認為憂鬱個體在與環境的互動中持續得到負向回饋是憂鬱持續的原因。依循認知及人際理論的脈絡，本研究嘗試探討負向的非語言訊息在憂鬱的人際認知整合模式中所扮演的角色，進一步探討社交／依賴型人格傾向與自主／自責型人格傾向的素質壓力模式。

研究的樣本是31位憂鬱症患者，以電腦化的作業評估針對臉部情緒表情刺激的負向認知基模反應及人際情緒經驗，以問卷評估人格傾向、尋求再保證後果及憂鬱狀態。

研究結果發現：（一）憂鬱症患者的負向認知基模反應與憂鬱嚴重度呈正相關，在負向及正向臉部情緒表情刺激時均有此一現象。（二）負向人際情緒經驗及他人的負向回饋在憂鬱衍生負向認知基模反應影響憂鬱的歷程中並無調節的效果，亦不經由憂鬱衍生負向認知基模反應的中介影響憂鬱狀態，但憂鬱衍生負向認知基模反應可經由正向人際情緒經驗的中介影響憂鬱狀態。（三）社交／依賴型人格傾向可經由負向人際情緒經驗的調節影響憂鬱，而自主／自責型人格傾向則經由憂鬱衍生負向認知基模反應的中介降低正向人際情緒經驗進而影響憂鬱狀態。

本文最後針對憂鬱衍生負向認知基模反應、臉部情緒表情刺激與憂鬱的關係及社交／依賴型人格傾向與自主／自責型人格傾向的素質壓力模式進行討論，並提出臨床應用上的可能性及進一步的研究方向。

Abstract

In the cognitive view of depression, depressogenic schemata activated by stressful events often causes an individual to become depressed and sustain depressive mood. Interpersonal model of depression postulates that negative feedbacks from others during the interpersonal interactions cause one to maintain depressed mood.

Accordingly, this study aimed to explore the functions of nonverbal information acquired based on the framework of integrated interpersonal-cognitive model of depression, and to examine the differences of diathesis stress paths between sociotropic/dependent and autonomous/self-critical types and depression.

Participants were 31 clinically depressed patients. Negative schemata activation toward facial emotional expressions was evaluated by a computerized program. Personality trait, reassurance seeking outcome and depression were evaluated by questionnaires.

The results showed that, first, negative schemata activation correlated positively with depression severity for both negative and positive facial emotions in depressed patients. Second, neither negative interpersonal emotion experience nor negative feedback from others moderated the relation between negative schemata activation and severity of depression. Negative schemata activation did not mediate either the path from negative interpersonal emotion experience or negative feedback from others to severity of depression. Third, negative interpersonal emotion experience had moderating effects on the path from sociotropic/dependent personality trait to depression. It is suggested that an individual with autonomous/self-critical personality trait may reduce positive interpersonal experience through activation of negative schemata and then become depressed.

Finally, this thesis discusses the relation among negative schemata activation, facial emotional expression and depression. More discussions are addressed on the diathesis stress model of sociotropic/dependent personality trait and autonomous/self-critical personality trait. According the results, suggestions for clinical practice and future researches are also provided.

經前症狀與生活壓力感受的關聯性：前瞻性逐日記錄之研究取向

The Relationship between Premenstrual Symptoms and Life Stress Perception: A Prospective Daily Rating Approach

摘要

本研究採用前瞻式逐日記錄的資料收集方式，探討婦女經前期之身心變化與其生活壓力感受的關聯。文獻資料顯示，5%至90%的婦女在經前期經驗嚴重程度不等之情緒低落、緊張焦慮、激躁不安、注意力無法集中、疲勞、食慾與睡眠改變，以及身體部位腫脹等症狀干擾。一般相信，荷爾蒙與神經內分泌物質的改變、人格特質、壓力因應能力，以及社會文化相關之態度與信念等多重變項，皆可能影響經前期的症狀表現。

本研究以 Rubinow (1989) 提出之「經前症候群是一種狀態調節的疾病」的概念架構為基礎，並假設：月經週期性的身心變化，使一些易感性較高的婦女產生知覺、認知和情緒等各方面之狀態依賴的變化；反映在其對生活壓力的評估明顯改變，在經前期感受到更大的壓力。

本研究受試者為一般生育期婦女，排除目前懷孕、哺育、手術或長期服用藥物等情況。受試者在42天記錄期間逐日填答「經前症狀量表」與「生活壓力事件量表」。本研究篩選之症狀組，須符合DSM-IV「經前期心情沮喪疾患」之診斷準則，且其症狀嚴重度在經前期明顯大於經後期30%以上。

本研究結果顯示：

1. 在開始逐日記錄的前幾天，經前症狀與生活壓力得分傾向於顯著高於之後的其他日子。

2. 在本研究年齡介於18至43歲的樣本中，受試者年齡越大，其經前症狀越嚴重。此外，受試者週期變動程度越大，其年齡越小、月經週期天數越長或是月經來潮持續天數越長。

3. 不同教育背景的受試者，其經前症狀與生活壓力略有差異。專科生（護專生）之「身體部位腫脹敏感」症狀顯著高於研究生或大學生；而大學生之「疲倦／性慾改變」則顯著高於專科生。此外，研究生與大學生之「成就表現」壓力顯著高於專科生。

4. 本研究估計經前症狀之盛行率為34.2%。

5. 症狀組或非症狀組在經前期皆有部分症狀增加的現象，然而，只有症狀組在經前期有明顯大幅提升之情緒症狀（煩躁不安、沮喪鬱悶和對人不友善）。

6. 症狀組在經前期感受到生活壓力增加，且以「一般人際衝突」的改變最為明顯。

最後，根據研究結果，本研究討論前瞻式逐日記錄之評量偏向效果、受試者特性的影響，以及「狀態依賴的變化」之假設概念在探討經前期變化的可行性，進一步提出未來的研究方向與可能的臨床應用。

Abstract

Previous studies revealed that 5 to 90% of women have more or less experienced premenstrual symptoms of low mood, tension, irritability, poor concentration, fatigues, changed appetite and sex drive, and physical discomforts. Hormonal fluctuations, per-sonality, stress-coping styles, and sociocultural factors are postulated to explain such a problem. As noted, psychological approach to Premenstrual Syndrome has been growing as an important area of research on women''s mental health.

In attempting to verify the link of menstrual cycle with daily stress perception as well as physical and psychological symptoms, the present study, based on the presumption suggested by Rubinow (1989) that premenstrual syndrome is a disorder of state regulation, was thus designed with a prospective, daily rating manner.

In this study, 117 menstruating women recruited from various settings in Taipei Metropolitan completed Premenstrual Symptom Scale (PSS) and Life Stress Scale (LSS) at a daily rating basis in 42 consecutive days. Of 117 women, 34.2% women (N = 40) were identified as PMS group and the rest as non-PMS group. The PMS group consists of those who not only met the DSM-IV diagnostic criteria of Premen-strual Dysphoric Disorder (DSM-IV) but also presented at least 30% more symptoms at premenstrual phase in comparison to postmenstrual phase. MANOVA and t-test were utilized in data analyses.

The results indicate that:

1. A significant drop of scores in daily rating approach was noted, in that scores of PSS and LSS were significantly greater in the first couples of days than of the following days;

2. Age is positively correlated with severity of premenstrual symptoms;

3. Education level plays a role in experience of premenstrual symptom (on PSS) and daily stress perception on LSS, in that students from nursing college reported more “body swelling”, undergraduates more “fatigue-changed sexual drive”, and graduate and undergraduate students more “achievement-related stress”;

4. In terms of the relationship between symptom manifestation and menstrual cycle, while both PMS and non-PMS groups reported significantly more PMS symptoms at premenstrual phase, only PMS group experienced more negative affects at premenstrual phase; and

5. In terms of the relationship between daily stress perception and menstrual cycle, only PMS reported significantly more stress in general at premenstrual phase, and more “non-specific interpersonal conflicts” in particular.

Results of the present study suggest a possible relationship among daily stress perception, premenstrual symptoms, and menstrual cycle. Utilization of prospective daily rating approach, screening procedure for potential PMS cases, and the concept of PMS as state-dependent changes are thus discussed. Implication for future research in this line and clinical applications are also suggested.

憂鬱病譜中個體的作業記憶功能缺損：
以情緒性語言與非語言刺激材料之N-Back作業表現為例
Working Memory Impairment of Individuals with Depressive Spectrum Disorders:
The Performances on N-Back Tasks Using Emotion-Valenced Verbal and Nonverbal Stimuli

摘要

回顧文獻發現，憂鬱個體的認知功能缺損，除了表現在認知建構上（如：態度或歸因等），尚涉及訊息處理速度、注意力、短期記憶、長期記憶、決策、及規劃等廣泛的訊息處理歷程，而造成廣泛性認知功能缺損的可能原因之一，即為憂鬱個體的作業記憶功能缺損或偏誤。現存文獻亦指出，影響作業記憶功能的調節變項包括研究樣本的特性及測量工具的特性，如：刺激材料的情緒性、作業費力的程度、刺激材料的語言或非語言屬性等。

奠基於憂鬱病譜（depressive spectrum）及情緒一致記憶（mood-congruent memory）的概念，並同時兼顧對於作業記憶的調節變項，本研究之主要目的在於以實徵方式探討憂鬱病譜中個體，面對情緒性語言和非語言刺激之作業記憶功能表現，並驗證高憂鬱者可能的作業記憶功能缺損或偏誤，以及憂鬱程度與作業記憶負荷量和刺激材料情緒性之間的關聯性。

本研究分為兩個子研究，研究一以憂鬱病譜中的大學生為樣本來源，依MINI確認診斷之有無，搭配BDI-II分數劃分為憂鬱組（n = 23）、情緒低落組（n = 25）、及控制組（n = 32），並控制其年齡、教育程度、語文智商、作業智商、及全量表智商等人口學變項與干擾變項。研究一為3 ×（3 × 3）的三因子混合設計，依序施測實驗一（語言刺激材料）和實驗二（非語言刺激材料）。組間獨變項為憂鬱程度（憂鬱組、情緒低落組、及控制組），組內獨變項為刺激材料情緒性（快樂、中性、及悲傷）與作業記憶負荷量（0-Back、1-Back、及2-Back），依變項為反應時間與正確率。研究二以臨床憂鬱患者為樣本（n = 18），並以MINI確認診斷，輔以BDI-II測量其憂鬱程度。研究設計為3（刺激材料情緒性）x 3（作業記憶負荷量）的二因子受試者內設計，組內獨變項及依變項皆與研究一相同。

研究一的分析結果，不論實驗一（語言刺激材料）或實驗二（非語言刺激材料），以正確率為依變項時，作業記憶負荷量和刺激材料情緒性的主要效果顯著，且憂鬱程度與作業記憶負荷量的交互作用顯著；以反應時間為依變項時，作業記憶負荷量的主要效果顯著，且憂鬱程度與作業記憶負荷量的交互作用顯著。研究二的分析結果，不論實驗一（語言刺激材料）或實驗二（非語言刺激材料），以正確率或反應時間為依變項時，都只有作業記憶負荷量的主要效果顯著。

綜合看來，不論以正確率或反應時間為依變項指標，作業記憶負荷量的主要效果皆顯著，故本研究採用N-Back作業確實能夠有效操弄作業記憶負荷量，且反應時間比正確率對於作業記憶負荷量的變化來得敏感。對於非臨床憂鬱個體而言，0-Back、1-Back、及2-Back 的情況皆非十分費力的 (effortful) 作業，僅在語言的 2-Back 作業上，憂鬱組的反應時間顯著較長。對於臨床憂鬱症患者而言，0-Back和1-Back亦非十分費力的作業，但到了 2-Back 已開始感到費力。此現象符合 Ellis 和 Ashbrook（1988）提出資源配置模式和認知干擾理論。本研究僅部份支持情緒一致記憶的效果，同時亦發現刺激材料的特性不同，可能導致不同的情緒不一致記憶效果。最後，探討本研究可能的貢獻與臨床應用，並提出研究可能的限制以及未來研究可以進行的方向。

Abstract

It has been often noted in literature that depressive individuals may suffer from broad deficits/bias in several areas of cognitive functions, which may be possibly due to the dysfunctions of working memory. Incorporated the possible moderators of working memory dysfunctions and/or impairments with the conception of depressive spectrum and mood-congruent memory, the main purpose of the present study was to investigate the impairment/bias of working memory of individuals with depressive spectrum disorders using N-Back task with verbal and nonverbal emotion-valenced stimuli.

Specifically, the present study examined: 1) the main effect of depression severity; and 2) the interaction effects between depression severity and working memory loadings as well as between depression severity and the valences of stimuli. The present study includes two sub-studies. Study-I recruited the students of the National Taiwan University as samples who were assigned to depressive group (n = 23), dysphoric group (n = 25), and normal control group (n = 32) based on the diagnosis of MINI and BDI-II scores. The demographic and other relevant variables such as age, education level, VIQ, PIQ, and FSIQ were controlled. The Study-I, with a 3 × (3 × 3) mixed design, consisted of N-Back experiment One (verbal version) and experiment Two (nonverbal version). In that, the within-group independent variables included valences of stimuli (positive, neutral, and negative) and working memory loadings (0-Back, 1-Back, and 2-Back). The dependent variables were accuracy and reaction time. Study-II recruited clinical depressive patients (n = 18) as samples. The Study-II, with a (3 × 3) within group design, consisted of N-Back experiment One (verbal version) and experiment Two (nonverbal version).

The results of Study-I, both verbal or nonverbal experiments, while indexed by accuracy, showed the main effects of working memory loadings and valences of stimuli, the interaction effects of working memory loadings with valences of stimuli and with depression severity. While indexed by reaction time, the results of Study-I revealed the main effect of working memory loadings and the interaction effects of working memory loadings with valences of stimuli and with depression severity. Differently, the results of Study-II showed the only main effect of working memory loadings in response to verbal or nonverbal stimuli.

The discussion of the present study, whether the dependent variables were accuracy or reaction time, there were significant main effects of working memory loadings, indicating of successful manipulation of the N-Back task for loading of working memory. For clinical patients, neither 0-Back nor 1-Back was effortful-enough task; however, 2-Back appeared to be an effortful-enough task. The present study partly supported the mood-congruent memory, but there were opposite trends about sad and happy stimuli in verbal and nonverbal experiments. Finally, we discussed the contributions and clinical applications of the present study, addressed the possible limitations, and provided some approaches to carry on further researches.

情緒性非語言行為的覺知特性與精神分裂病質傾向的關連性探討

Nonverbal Emotional Perception and Schizotaxic Tendency: A Comparative Study of Schizophrenic Patients, their families, and the Normal Control

摘要

過去許多研究指出，精神分裂症患者在臉部表情與聲音情緒的辨識會出現困難；亦有部分研究顯示精神分裂症患者在其他非語言溝通的辨識上亦有困難。然而，這些研究多半侷限於單一管道（如臉部表情/肢體動作/聲音）的現象探討，探討的對象也侷限於精神分裂症的患者。本研究的主要目的，在於引用Meehl與Tsuang兩位學者所提的精神分裂病質（schizotaxia）之觀念並加以擴充，研究上將對象擴充至精神分裂症的一等親屬，探討他們在非語言情緒的辨識上是否與患者或一般人有所差異，並試圖運用非語言情緒辨識中不同管道（臉部表情及聲音）間的不一致性來瞭解精神分裂病質的可能表現。

本研究的進行，分為前置性非語言情緒覺知工具的編製研究與正式的組間比較研究。在前置性非語言情緒覺知工具的編製研究中，本研究以情緒誘發方式拍攝或錄製模特兒的表情與聲音，經過常模評分歷程，最後選取情緒類屬與強度同意度適中，且符合研究理念的表情照片30張與語音情緒30段，作為台灣版非語言覺知診斷性分析工具（DANVA2-Taiwan version）的刺激材料。所有的刺激材料均轉製成電腦版本，完全以手提電腦呈現標準化刺激，並紀錄反應結果。在正式研究中，本研究的研究對象包括精神分裂症患者二十六名，精神分裂症患者的一等親屬中沒有重大精神疾患及身體不適者二十五名，以及家族中沒有明顯精神分裂病史的正常人三十九名。研究工具包括台灣版非語言覺知診斷性分析工具中的單管道表情、單管道語音等兩個分測驗，雙管道情緒一致性與不一致性辨別測驗，以及魏氏成人智力測驗第三版中的五個分測驗。資料的收集採個別施測方式，分別測量三組受試者對於臉部表情及聲音情緒的辨識能力，以及受試者在不同管道傳遞一致與不一致情緒時受試者的辨識能力，並施以結構式訪談與自填式症狀量表。

本研究的資料經相關分析、變異數分析、t-檢定、與迴歸分析後，結果發現：對於非語言情緒性刺激的理解或辨認，在精神分裂症患者、家屬、及一般人間有顯著並且方向一致之差異，此一差異組型分別出現在臉部表情及聲音的管道上。

精神分裂症患者與一等親屬在臉部表情與聲音情緒的非語言情緒辨識上，有介於一般人與患者之間的輕微偏差，而以聲音管道的偏差較為明顯。本研究進一步檢視各個情緒辨識能力，發現家屬與患者在非語言憤怒情緒的辨識與一般人有明顯差異，而此一差異涵蓋表情及聲音兩個管道。

根據分析結果，在不同的管道之間同時給予不一致的刺激時，一般人選擇傾向先選擇聲音作為主要判斷依據，且以臉部表情為主或以聲音為主的正確率大致相同，但是精神分裂症的患者及其家屬以表情為主要判斷線索的比例顯著較以聲音為主的比例為高。對於一致或不一致的非語言情緒的整體覺知能力顯著較一般人為差，而一致性的非語言訊息對於一般人、家屬以及患者有不同程度的輔助效果。本研究認為非語言情緒辨識能力在精神分裂症患者與家屬與正常人之間呈現連續性的差異，有可能與精神分裂病質傾向的荷重相關。因此，本研究的研究工具、測量方法與發現，對於後續有關精神分裂病質觀點的研究探討，以及研擬針對精神分裂病患與家屬的非語言情緒特性之篩檢與矯正之臨床應用，或可提供有益的參考。

Abstract

Studies in the past suggest that the schizophrenic patients have difficulties in recognizing emotions within facial expressions, paralanguages, and other nonverbal channels. However, most of these studies were limited in examining only one channel (i.e. facial expressions/paralanguages/postures and gestures), and in using only schizophrenic patients, rather than simultaneously examining two or more channels and including their close relatives as subjects. Based on and expanding the concept of schizotaxia introduced by Meehl and Tsuang, the present study compared the nonverbal emotion recognition abilities within different channels among schizophrenic patients, their close relatives and normal subjects. The present study also investigated the differences among three groups of subjects while given simultaneous congruent or incongruent emotional cues of different nonverbal channels.

The present study was proceeded with a pilot study beforehand to establish an effective tool for evaluating nonverbal accuracy, and then followed by formal study comparing the differences among schizophrenic patients, their close relatives and normal controls. In the pilot study, we photographed pictures of facial expressions and recorded emotional voice clips after mood-induction tasks, and finally chose 30 applicable facial expression photographs and 30 emotional voice sections. All of these stimuli were highly agreed by an independent sample of subjects from campus. These emotional stimuli were used to form the Diagnostic Analysis of Nonverbal Accuracy 2-Taiwan version (DANVA2-TW). All chosen photographs and voice sections were transformed to a computerized version and presented to each subject in a random order. Their responses were also recorded with notebook computers. In the formal study, our subjects included 26 schizophrenic patients, 25 first-degree relatives of schizophrenic patients without any major psychiatric diagnoses, and 39 normal controls without any obvious psychiatric history. Instruments used in this study include Single-channel Facial and Single-channel Paralanguage from DANVA2-TW, Bi-channel congruent & incongruent nonverbal emotional discrimination tool (which was derived from DANVA2-TW), and five subtests from WAIS-III, including the Digit Span, the Block Design, the Arithmetic, the Digit Symbol Substitution, and the Information. Structured interview and self-report symptom checking list were also given to each subject. Each subjects was examined individually with standardized procedure at hospital or home settings.

Data were analyzed by correlation analyses, analyses of variance, t-tests and regression analyses. The results showed significant differences of nonverbal emotion recognition abilities among schizophrenic patients, their first-degree relatives, and normal controls in both facial and paralanguage channels. More specifically, both patients and their relatives displayed less accurate performance than the normal controls, and the performances of relatives were as worse as patients’ or the differences were smaller in the relatives than in the patients. In terms of the differences between various channels, the differences were much more significant in the paralanguage channel. The present study further examined the recognition abilities of various emotions, and found that patients and their relatives performed significantly worse than normal subjects in recognizing anger emotions within both facial expression and paralanguage channels.

According to the results, while incongruent emotions were delivered from facial expressions and paralanguages, normal subjects tend to choose facial and paralanguage channels as the major one at an equal rate; and they could find out both emotions delivered from facial expressions and paralanguages correctly. However, schizophrenic patients and their close relatives tend to choose facial expressions as their major channel for decision at the first moment, and their accuracy for correct emotions, both of congruent and incongruent cues delivered from different channels, was significantly worse than the normal controls. Congruent emotion cues from different nonverbal channels were found to have different enhancing effects on the accuracy of nonverbal emotion perception among schizophrenic patients, close relatives and normal controls.

Finally, the present study suggests that, in the aspect of nonverbal emotion recognition ability, there is an increasing trend among schizophrenic patients, close relatives and normal controls, and the differences of the ability may be probably related to genetic loading of schizotaxic tendency. The present study not only provides feasible instruments for measuring nonverbal accuracy, but also explores practicable methods for future research on schizotaxia and schizotaxic trait. Moreover, the findings of the present study also suggest further investigating schizotaxia from the viewpoint of nonverbal communication, especially the aspect of nonverbal emotional perception.

提升性歸因與憂鬱的復原之前瞻性研究
A Prospective Study of Enhancing Attributions and Recovery from Depression

摘要

本研究主要的目的在於驗證 Needles 與 Abramson（1990）所提出的「憂鬱復原模式」，並試圖探究情緒低落的個體自發性的緩解機制。「無望憂鬱理論」（Abramson, Metalsky, & Alloy, 1989）指出，具有憂鬱衍生性歸因風格的個體，面臨負向生活事件時，傾向於將負向生活事件歸因為穩定與概括的因素，因而產生無望，並導致憂鬱的症狀。相對的，「憂鬱復原模式」主張，具有提升性歸因風格（亦即，將正向事件歸因為穩定與概括因素之傾向）的憂鬱個體，在遇到正向生活事件時，其交互作用會導致希望的回復，並使其自憂鬱之中復原。然而，過去相關的實徵研究在方法與結果上，仍有諸多歧異。驗證「憂鬱復原模式」之實徵研究中，皆是以「無望的下降」來代表「希望的回復」；但是，希望與無望不見得是單一向度下的兩個端點。再者，過去的研究以真實或假想的情境來測量歸因型式，所獲得的研究結果不盡相同。此外，究竟是提升性歸因風格、正向生活事件，抑或這兩者之間的交互作用之效果，預測了希望的回升與憂鬱症狀的下降，在過去的研究中亦未得到一致性的答案。

針對上述之歧異，本研究試圖以本地情緒低落的大學生為樣本來進行檢驗，並提出三個研究假設：（一）對正向生活事件的提升性歸因風格與正向生活事件的交互作用，能預測對新近正向生活事件的提升性狀態歸因。（二）對新近正向生活事件的提升性狀態歸因，能預測希望的提升。（三）希望的提升，能預測無望憂鬱症狀的減少。

研究一首先以 265 位大學生為樣本，檢驗本研究所編製的中文版狀態希望量表之心理計量特性；研究二以 61 個情緒低落的大學生為研究參與者，採用前瞻性縱貫式的研究法，以三週的時距，重複四次測量研究參與者的狀態希望、憂鬱情緒、狀態歸因、以及情緒上揚事件，並在第一週測量研究參與者的歸因風格。研究結果顯示，提升性狀態歸因與情緒上揚事件合併的主要效果，能預測希望的上升；並且，希望的上升以及提升性狀態歸因與情緒上揚事件之間的交互作用效果，皆能預測無望憂鬱症狀的下降，此研究結果大致上符合本研究之假設與「憂鬱復原模式」。不過，與研究假設不一致之處在於，對先前正向生活事件之內在與概括歸因風格，反而預測無望憂鬱症狀增加，此結果有可能導因於研究參與者的特性或文化差異之影響所致。

最後，針對本研究的貢獻與臨床上可能的應用進行討論，並提出研究可能需要改進的地方，以及提供未來研究可以進行的方向。

Abstract

The main purpose of the present study was to investigate the Model of Recovery from Depression proposed by Needles and Abramson (1990), and to attempt discovering the mechanism of spontaneous remission of dysphoric individuals. Hopelessness Theory of Depression (Abramson, Metalsky, & Alloy, 1989) indicated that when negative life events occurred, individuals who had depressogenic attributional style tended to attribute them to stable and global causes which would thus produce hopelessness and lead to symptoms of depression. On the contrary, the Model of Recovery from Depression postulated that when depressed individuals with enhancing attributional style（i.e., the tendency to attribute positive events to stable and global causes）were confronted with positive life events, the interaction between them, would initiate the restoration of hopefulness, and then lead to the recovery from depression. However, the methodologies and results of the previous related empirical studies were inconsistent. All of the empirical studies which examined the Model of Recovery from Depression used “reduction of hopelessness” to represent the “restoration of hopefulness”. However, hopefulness and hopelessness may not be the two ends of the same dimension. Moreover, previous studies used either real or hypothetical situations to test attributions and let to inconsistent conclusions. Furthermore, prior studies which examined the enhancing attributional style, positive life events, or the interaction between these two variables to predict the restoration of hopefulness and the reduction of depressive symptoms had not yet received the consensus.

In consideration of the above inconsistencies, this study attempted to investigate the problems with dysphoric undergraduates by examining three hypotheses: (1) The interaction of positive life events and enhancing attributional style for positive life events will predict enhancing state attributions. (2) Enhancing state attributions will predict increase of hopefulness. (3) Increase of hopefulness will predict decrease in symptoms of hopelessness depression.

Study one tested psychometric characteristics of the Chinese State Hope Scale developed in this study with a sample of 265 undergraduates. Study two recruited 61 dysphoric undergraduates and adopted the prospective longitudinal design, in which the state hope, depressive mood, state attributions, and uplifts were repeatedly measured once a week for four times within a duration of three weeks. At Time 1, we also assessed their attributional style. The results showed that the combined main effects of enhancing state attributions and uplifts predicted increase of hopefulness. In addition, both the increase of hopefulness and interaction of enhancing state attributions and uplifts predicted decrease in symptoms of hopelessness depression. These results generally met the hypotheses of the present study and the Model of Recovery from Depression. Nevertheless, contradictory to our hypotheses, internal and global attributional style for the previous positive life events contrarily predicted increase of symptoms of hopelessness depression. The result may be due to the diversities of the study methods, the characteristics of the participants, or the influences of cultural differences.

Finally, we discussed the contributions and the clinical applications of the present study, addressed the possible limitations of the study, and provided some directions to carry on further researches.

情緒低落者注意力現象初探：行為抑制與行為激發系統的調節角色

Exploring Attention of Individuals with Depressed Mood: The Moderating Roles of Behavioral Inhibition and Activation System

摘要

根據Beck (1967)的基模理論、Bower (1981) 的網絡聯結理論以及Clark與Watson (1991) 的三元模式，可推論憂鬱者憂鬱心情與負向訊息處理偏誤有關，而正向訊息處理偏誤缺乏正向情緒有關。過去憂鬱者注意力投注與注意力脫離現象的相關研究結果，顯示雖具有負向及正向訊息注意力偏誤的傾向，但較不穩定。因此本研究嘗試以Gray (1982) 的行為抑制及行為激發系統做為第三變項，探討其對憂鬱者注意力現象特徵的調節作用，並同時將注意力現象特徵與情緒反應加以結合，期待此種整合性地研究設計，可以更進一步檢驗過去理論所提出的情緒一致性假說。

本研究共包含兩大研究，研究一為發展本地適用的中文版行為抑制與行為激發量表，第一階段以150位大學生為樣本進行因素分析，以此分析結果做為題目潤飾的基礎，第二階段則另招募152位大學生進行潤飾後中文版量表因素結構及心理計量特性檢驗。研究二則以119位大學生為樣本，檢驗憂鬱傾向與注意力現象之關聯是否會受到行為抑制與行為激發系統活化程度的調節，並探討注意力現象對情緒反應的影響。

本研究結果顯示（1）中文版行為抑制與行為激發量表為一具有良好心理計量特性，可有效測量相關概念的本土化工具。（2）整體來說，憂鬱程度與負向及正向刺激的注意力現象特徵無顯著相關，但憂鬱傾向與負向注意力投注之關聯會受到行為抑制系統活化程度的調節，行為抑制系統活化程度較低時，憂鬱傾向與負向注意力投注有顯著的負相關。（3）注意力現象與情緒誘發前後情緒改變量無顯著關聯。本研究嘗試以迴向抑制效果及警覺-逃避理論之角度切入，討論行為抑制系統調節效果的可能意涵，並進一步說明研究的可能貢獻與限制，以及未來可行的研究方向。

Abstract

According to the Schema theory (Beck, 1967), Associative network theory (Bower, 1981) and Tripartite theory (Clark & White, 1991), previous researchers suggested the correlation between negative attention bias and depressed mood, as well as positive attention bias and the lack of positive mood. However, empirical studies did not reveal consistent findings on attentional phenomena. Therefore, the present study aimed firstly to investigate the moderating role of the behavioral inhibition and activation system (Gray, 1982) between attentional engagement and disengagement in depressors, and secondly, to examine the relationship between individuals’ attentional phenomena and depressed mood.

With an integrated design, the present study intended to further examine the mood consistent hypotheses. Two studies were included. Study one aimed to validate the Chinese version of the Behavioral Inhibition and Behavioral Activation Scale（BIS/BAS）. In the first stage (N = 150), factor analysis was performed, and scale items were revised accordingly. In the second stage (N = 152), factor analysis was performed on the revised scale, and its psychometric properties were examined. Study two (N = 119) aimed to examine: 1) whether the relationships between depressed mood and attentional biases were moderated by the activation level of behavioral inhibition system (BIS) and behavioral activation system (BAS); and 2) the relationships between attentional biases and mood changes after mood induction.

Results indicated that, first of all, the BIS/BAS scale showed satisfactory psychometric properties. Secondly, the correlations of depressed mood and attentional biased were insignificant overall. However, the moderating role of the BIS between depressed mood and negative attentional engagement was revealed. Specifically, when the BIS was less active, significant negative correlation was noted between depressed mood and negative attentional engagement. Finally, the correlations between participants’ attentional bias and the mood changes before and after mood induction were insignificant. Based on the above findings, the present study tried to use structure aspects such like Inhibition of Return (IOR) and Vigilance-Avoidance Theory to discuss the possible meaning of moderating effect. Possible application and future research directions were also addressed.

九二一震災後災區青少年之創傷後壓力反應：性別差異及創傷暴露效果分析

Posttraumatic Stress Reaction of Adolescents After the Chi-Chi Earthquake: Gender and Exposure Effect Analysis

摘要

本研究之目的在瞭解九二一地震後，災區國中青少年的創傷後身心反應，包括創傷後壓力症狀及身心症狀。並試圖以一概念模式整合創傷暴露、基本人口變項、性格特質以及個人脆弱性對青少年創傷後身心反應之影響力。

在九二一地震發生將近一年後，對1173名東勢及埔里兩災區之國中青少年症狀。其中，自己受傷、好朋友受傷或死亡、鄰居受傷、主觀知覺地震威脅程度以及地震後與父母分開居住等創傷暴露變項明顯影響創傷後壓力症狀及身心症狀之嚴重度。進行調查，樣本年齡分佈為13-15歲。使用工具為自填式問卷，包括：自編基本資料及地震損傷問卷、UCLA PTSD INDEX FOR DSM-IV（青少年版）、身心症狀量尺及自我效能量尺。

本研究結果顯示：

1. 九二一地震一年後，災區國中生仍受到地震的影響，持續出現創傷後壓力症狀及身心症狀。

2. 地震中受創程度越嚴重的青少年，報告越多創傷後壓力症狀及身心。

3. 不同性別之青少年在症狀嚴重度及常見症狀表現上有所差異。整體而言，女性較男性呈現較嚴重的創傷後壓力症狀及身心症狀。相較於男性，女性報告較多重複經驗及逃避症狀；但男女在心理麻木及其他適應不良症狀的表現則無明顯差異。

4. 創傷暴露程度、年齡、性別、過去創傷經驗、以及自我效能感分別能預測23%的創傷後壓力症狀變異量，以及25%的身心症狀變異量。

5. 過去創傷經驗及自我效能對創傷後身心反應之預測力高於創傷暴露程度之預測力。

最後，研究者提供未來可行的研究方向及本研究結果在臨床應用上之意涵。

Abstract

This study investigated post-traumatic stress reactions in adolescents (13 to 15years) from two heavily impacted townships in Central Taiwan near one year after the Taiwan Chi-Chi Earthquake. An integrative conceptual model, containing 4 factors, i.e., exposure level, personal characteristics, personality and personal vulnerability, was used to examine the contribution of multiple risk factors in predicting post-traumatic stress reactions.

One year after the earthquake, 1,173 adolescents were assessed using the modified Earthquake Exposure Index, UCLA PTSD INDEX for DSM-IV (adolescents version), Psychosomatic Scale, and Self Efficacy Scale.

The results reveal that one year after the Chi-Chi earthquake, adolescents from Central Taiwan continuingly reported some PTSD and psychosomatic symptoms. Those who experienced higher level of exposure reported more post-traumatic stress reactions. Personal injury, perception of threat, and separation with parents after the Earthquake are important exposure variables, and accounted for significant portion of variances of PTSD and psychosomatic symptoms. At the symptom level, female reported more re-experience and avoidance symptoms than males, but male reported no less numbness symptoms. Multiple regression analyses show that exposure, gender, age, self-efficacy, and previous traumatic experiences accounted for 23% and 25% of the variances of PTSD and psychosomatic symptom, respectively. Self-efficacy and previous traumatic experiences are more powerful than exposure in predicting chronic post-traumatic stress reactions.

Based on the findings, constituents of PTSD, nature of the disasters, exposure and gender effects, research designs and methodological considerations, as well as future research in the line of trauma study for adolescents were discussed. Clinical applications were also addressed.

社交焦慮、網路社交焦慮與網路環境特性之關聯性探討
The Relations of Social Anxiety, Internet Social Anxiety
and Characteristics of the Internet

摘要

本研究之目的在於瞭解社交焦慮的本質與社交焦慮者的網路人際互動現象，並試圖以社交焦慮的認知行為特性，以及網路純文字交談環境的特性，來探討高社交焦慮者的網路使用狀況。本研究進一步瞭解網路人際互動提供高社交焦慮者何種層面的優勢，以及此優勢與網路社交焦慮特性之間的關聯。

本研究運用Mattick 和 Clarke (1998) 所編製的社交互動焦慮量表及社交恐懼量表、自編的網路使用狀況調查問卷、網路社交焦慮量表，以及網路人際互動與面對面人際互動比較問卷，對237名平均年齡20.68歲的受試者施測。本研究除採用相關及迴歸分析之外，並依據中文化後的社交互動焦慮量表及社交恐懼量表，將受試者區分為高社交焦慮組及一般組進行比較，得到的結果顯示：

1. 一般情境下的社交焦慮與網路上的社交焦慮有顯著正相關。

2. 高社交焦慮傾向者的網路人際互動或需求並不比一般人高。網路人際互動或需求也未與網路社交焦慮、網路評價焦慮有關，但和網路交談焦慮呈顯著負相關。

3. 不同類型的社交焦慮對整體的或不同類型的網路社交焦慮有不同的影響。其中，表現焦慮對整體的網路社交焦慮之預測性大於互動焦慮；互動焦慮預測網路交談焦慮，而表現焦慮則預測網路評價焦慮。

4. 高社交焦慮者比一般人感受到更多的網路人際互動的優勢，包括較能自在表達、表現平時壓抑的特質、減少了面對面的焦慮，也能因此滿足人際需求，並在與他人面對面互動時變得更自在，更有自信。

上述結果意謂了網路環境雖對社交焦慮有某種程度的助益，但整體而言，社交焦慮仍是較穩定的特質，受網路環境特性的影響有一定的限度。本研究進一步探討社交焦慮與網路特性之間的關聯對網路使用者及其行為的影響之研究價值與限制，並說明未來可行的研究方向與臨床應用。

AbstractThe purpose of this study was to understand the nature of social anxiety and the interpersonal phenomenon of socially anxious people in the Internet. From the framework of both cognitive-behavior model of social anxiety and characteristics of text-based conversation environment of the Internet, the study also aimed to find out the on-line patterns of users with high social anxiety, to explore the advantages of Internet interaction over face-to-face interaction, and to examine the relations between such advantages and social anxiety.

This study administered the Internet Usage Questionnaire, Internet Social Anxiety Scale, and Survey of Internet and Face-to-Face Interaction Comparison that were all developed for this study, along with the Social Interaction Anxiety Scale (SIAS) and Social Phobia Scale (SPS) (Mattick & Clarke, 1998) to 237 subjects (mean age 20.68）. Based on the scores of the Chinese version of SIAS and SPS, subjects were sorted into socially anxious and normal groups. In addition to correlation and regression analyses, further group differences were examined.

The results showed that:

1. Social anxiety correlated positively to Internet social anxiety.

2. Socially anxious group showed no more demands of interpersonal interaction in the Internet, as compared with the normal control. Internet interpersonal interaction correlated neither to Internet social anxiety nor to Internet evaluative anxiety, while it was noted to have negative correlation with Internet conversation anxiety.

3. Different subtypes of social anxiety had unequal influence on different subtypes of Internet social anxiety. Performance anxiety predicted Internet social anxiety better than interaction anxiety did. While interaction anxiety predicted Internet conversion anxiety, performance anxiety predicted Internet evaluative anxiety.

4. Socially anxious group reported to gain more advantages of interpersonal interaction in the Internet than the normal control, including that they could express themselves more easily, show repressed aspects of the self, suffer less social anxiety, satisfy their social needs, and become more relaxed and self-confident in face-to-face interaction.

In conclusion, although Internet did benefit socially anxious people, social anxiety, as a relatively stable trait, seems to be limitedly influenced by Internet characteristics. This study also discussed the values and limitations of study of the relations of social anxiety and characteristics of the Internet as well as their effects on Internet usages. Finally, suggestions for further research and clinical application were provided.

反芻型反應風格、自傳式記憶與憂鬱之關聯性探討
Ruminative Response Style and Autobiographical Memory:
An Integrative Correlational Model of Depression

摘要

Nolen-Hoeksema (1991) 之憂鬱反應風格理論提出，若個體傾向穩定地採取反芻型反應因應其憂鬱狀態，則將經驗到憂鬱狀態的持續維持或惡化。相關與實驗取向之實徵研究均已充分驗證反芻型反應風格對憂鬱之延長與惡化效果。在此基礎上，後續許多關於反芻型反應風格作用機制的探討，均指向個人記憶功能在反芻型反應風格與憂鬱狀態之間，扮演重要角色。鑑於本地探討類似議題的相關研究有限，可茲使用的本土化研究工具亦不足。因此，本研究目的在於發展合宜之研究工具與程序，藉以釐清反芻型反應風格與自傳式記憶功能之關聯性，以及兩者共同預測憂鬱狀態之路徑模式。

本研究共包括四個子研究，研究一以350名大學生為樣本，檢驗本研究所編製的中文反應風格量表之分量表結構與心理計量特性；研究二以158名大學生為樣本，驗證憂鬱反應風格理論，並提供中文反應風格量表之效度基礎；研究三發展文本取向自傳式記憶研究方法，並編製結構化的內容分析工具，以65名大學生為樣本，檢驗自傳式記憶過度概化特性與憂鬱狀態之關聯性；研究四則以 94 名門診憂鬱患者為研究對象，同時針對反芻型反應風格、過度概化記憶、及患者之憂鬱狀態，以路徑分析探索其互動預測模式。

本研究結果顯示：

1. 中文反應風格量表具良好之心理計量特性，為合宜之概念測量工具。

2. 反芻型反應風格與憂鬱症狀及主觀憂鬱心情均呈顯著正向關聯性。

3. 自傳式記憶過度概化程度與憂鬱狀態具顯著正向關聯性。

4. 反應風格與過度概化記憶共同預測解釋患者憂鬱症狀的總變異量之54.7%，以及主觀憂鬱心情的總變異量之25.1%。

5. 路徑分析結果顯示，反芻型反應風格可直接對患者的憂鬱症狀產生影響，並藉由過度概化記憶的部分中介，進而影響患者的憂鬱症狀。

最後，本研究由注意力資源與情緒記憶系統之角度切入，討論在憂鬱的維持機制中，反芻型反應風格對於個人認知資源與記憶功能的可能影響。此外，本研究報告亦討論未來可行的研究方向以及研究結果在臨床應用上的意涵。

AbstractThe Response Style Theory of Depression（Nolen-Hoeksema, 1991）suggests that individuals who respond to their depression with consistently engaging in rumination would exacerbate and prolong their depressive mood and symptoms. Both correlational and experimental studies have empirically verified the role of ruminative response style in the maintenance and exacerbation of depression. Based on extant empirical findings, recent studies have begun to explore the mechanism underlying rumination and its effects on the development of depression. Growing evidences suggest personal memory plays a critical role in the link between ruminative response style and depression. Yet, research of this line is to be done and relevant research tool is to be developed in Taiwan. Hence, the aim of this study is to develop essential research tool and appropriate methodology to investigate ruminative responses of depression. Moreover, the present study will further investigate the association between autobiographical memory and ruminative response style as well as their predictive ability for depression.

The present thesis consists of four studies. Study 1 examined the psychometric properties of the Chinese version of the Response Style Questionnaire (CRSQ) by using a sample of 350 undergraduate students. Study 2 examined the Response Style Theory of Depression by using a sample of 158 undergraduate students and provided validity evidence for the CRSQ. In Study 3, the author devised a modified context-based autobiographical memory task and a structured rating scale to encode memory content for further quantitative analyses. An examination of the association of overgeneral autobiographical memory with current depressive mood and symptoms was then conducted with a sample of 65 undergraduate students. In Study 4, the sample consisted of 94 depressive outpatients recruited from the department of psychiatry of a medical center. We examined the relationship among ruminative response style, overgeneral autobiographical memory, and current depressive symptoms and mood. Furthermore, we investigated the integrative correlational model of rumination, overgeneral memory and depression by path analysis.

Results revealed that:

1. The Chinese Version of Response Style Questionnaire yielded adequate and satisfactory psychometric properties, supporting itself as a useful measurement of response style of depression.

2. Ruminative response style significantly and positively correlated with depressive symptoms and current depressive mood.

3. Overgeneral autobiographical memory significantly and positively correlated with depressive symptoms.

4. Ruminative response style and overgeneral memory together accounted for 54.7% of the total variances in patients’ depressive symptoms and 25.1% of the total variances in patients’ current depressive mood.

5. Path analysis showed that ruminative response style affected patients’ depressive symptoms directly and was partially mediated by overgeneral memory.

Based on the findings, further discussion was provided to centre around the potential effects of ruminative response style on individual’s cognitive resources and memory function specifically in maintenance and exacerbation of depression. Clinical applications as well as future directions were also addressed.

創傷後壓力疾患個體對情緒訊息的自動化情緒處理：以人際創傷事件倖存者為例
Automatic Emotional Information Processing in Interpersonal Trauma Survivors with Posttraumatic Stress Disorder (PTSD)

摘要根據過去創傷後壓力疾患（posttraumatic stress disorder, PTSD）的心理病理模式，創傷記憶本質上的個別差異是影響疾病是否發展與維持的重要因子。在認知取向的心理病理模式中多半主張，個體具有兩種不同本質的記憶問題，一種是難以主動提取對創傷事件的完整記憶；另一種是非自主但會被自動激發的侵入式記憶，時常會以鮮活的多感官形式伴隨強烈情緒出現。後者被認為與創傷記憶之自動化激發有密切關聯，而創傷記憶被激發後，對後續行為、認知與情緒帶來之影響，也可能是造成創傷後壓力症狀維持的主要原因。以上現象皆涉及對情緒訊息的自動化處理歷程，然而，在創傷後壓力疾患的研究領域中，關於此歷程之機制目前仍未有定論，研究結果也不一致，仍需要更多研究來對其探討。本研究主要目的在於探索發展出創傷後壓力疾患之人際創傷倖存者對情緒性臉部表情的自動化處理，並且嘗試連結此自動化處理歷程與臨床症狀，期能提供後續心理專業介入之參考。

本研究參與者來源為大學部/研究所學生，依照其創傷經驗與創傷後壓力症狀嚴重度分為三組，分別為人際創傷PTSD組（18人）、人際創傷控制組（46人）與健康控制組（38人）。本研究採用情緒促發作業進行，刺激材料為彩色的人類臉部表情圖片，共有生氣、快樂與中性三種情緒，分別作為促發刺激與目標刺激。促發刺激與目標刺激之呈現間隔時間為200毫秒。研究程序方面，將會先進行創傷經驗、臨床症狀與相關特質之測量，接著完成以電腦施測的情緒促發作業與刺激材料威脅評估作業，最後進行簡短晤談。本研究假設在情緒促發作業中，人際創傷PTSD組將會和兩控制組展現出不同的反應傾向，而此組間差異僅會呈現在以生氣表情為促發刺激之情況下。

研究結果發現：（1）人際創傷PTSD組在情緒促發作業之表現明顯與兩控制組不同；（2）在情緒促發作業中，人際創傷PTSD組在生氣—快樂配對下的妨礙效果顯著小於其他兩控制組，其他配對情況則無任何顯著組間差異；（3）人際創傷倖存者在情緒促發作業上之表現和PTSD症狀嚴重度的相關達顯著。本研究結果支持人際創傷PTSD個體在自動化情緒訊息處理上，和一般個體具有不同傾向。當外界出現創傷相關訊息使PTSD個體的創傷記憶被激發後，可能會促使他們產生逃避處理創傷訊息之傾向，推論此自動化處理傾向或許為PTSD中情感麻木症狀的可能機制。本研究將根據上述結果提出臨床應用建議與後續研究方向。

AbstractAccording to past psychopathological models of posttraumatic stress disorder (PTSD), the individual differences of trauma memory may impact the development and maintain of PTSD. Based on cognitive approach to the psychopathology of PTSD, trauma memory may reveal two different kinds of problems. One is difficulty in intentionally retrieving a complete memory of the traumatic event; another is high frequency of involuntary triggered intrusive memories in a very vivid and emotional way. The later related to cue-driven process and have been commonly linked to automatic information processing. When trauma memory was automatically activated by trauma-related cues, it can have impact on subsequent behavioral, cognitive or emotional processes. That involves the automatic emotional information processing and may be one probable interpretation of why posttraumatic stress symptoms maintain. However, only few studies have investigated automatic emotional information processing in trauma survivors with PTSD but the findings were still inconsistent. Therefore, more studies focus on this topic is needed.

This study aimed to explore the automatic information processing toward emotional facial expression in interpersonal trauma survivors with PTSD and to investigate the relationship between this automatic process and clinical symptoms. Recruited undergraduate students were divided into three groups depend on their traumatic experience and the intensity of posttraumatic stress symptoms: interpersonal trauma survivors with PTSD (PTSD group, n = 18), interpersonal trauma survivors without PTSD (C_IPT group, n = 46), and healthy control group without interpersonal trauma history (C_H group, n = 38). Affective priming paradigm was used, with colorful photos of angry, happy, and neutral human faces as priming and target materials. Stimulus onset asynchrony between prime and target was 200ms. In experimental procedure, all participants will complete the measurements of trauma experience, clinical symptoms, and other related trait first. Then finish computerized affective priming task and material threatening evaluation task. After all, a brief semi-structured diagnostic interview was delivered. It is hypothesized that PTSD group would show different response tendency to C_IPT and C_H group. This effect only appears when angry face as prime stimulus.

Results revealed: (1) The performance of PTSD group in affective priming task was different from the other two control groups; (2) PTSD group showed less interference effect than the other two groups when the angry prime was followed by happy target; (3) The performance of interpersonal trauma survivors in affective priming task was significant related to their posttraumatic stress symptoms. These results suggested that, when trauma memory was activated by trauma-related stimulus, PTSD individuals may automatically try to minimize the impacts of such stimulus. The findings suggest that automatic activation of trauma memory has more impacts on consequent evaluative process in interpersonal trauma survivors inclined to PTSD psychopathology, such as emotional numbing. Based on these results, implication for clinical intervention and future research are discussed.

親友對憂鬱病患的反應：覺知尋求再保證與關係義務感的機制探討
The Responses of Relatives and Friends toward Depressed Individuals: The Mechanism of Perceived Reassurance-Seeking and Sense of Relation Obligation

摘要本研究旨在探究親友的覺知尋求再保證與關係義務感如何影響其對憂鬱者之反應。Coyne（1976b）提出憂鬱人際模式以描繪一個透過與周圍他人的互動進而影響憂鬱之維持或惡化的循環系統。在此系統中，憂鬱者的症狀會使他人產生負向情緒與拒絕行為，而他人的反應又會返身增強/維持個體的憂鬱。在此概念脈絡下，Joiner及其同僚自1992年開始，透過一系列的研究，進一步提出病患的「尋求再保證」乃是憂鬱人際理論的核心。截至目前，西方憂鬱人際理論的概念建構及實徵研究均有相當的進展，相較之下，台灣本地在此方面的探討卻仍屬闕如。有鑑於此，本研究整合西方的理論概念模式，並考慮台灣本地的文化特性，一方面檢驗西方憂鬱人際理論之預測在台灣本地臨床憂鬱樣本的適用性，另一方面主張「關係義務感」及「拒絕概念多面向性」乃西方理論所忽略之重要觀點，並進一步設計實徵研究來修正與拓展憂鬱人際理論。

本研究以九十四名門診憂鬱患者之親友為研究對象，透過自行發展的覺知尋求再保證與關係義務感等量表與訪談，結果發現：（一）親友覺知憂鬱者的尋求再保證與其負向情緒確為拒絕反應之影響因子，但台灣與西方所呈現的拒絕樣貌相當不同；（二）親友的關係義務感調節其覺知尋求再保證對拒絕及負向情緒的預測效果。是故，憂鬱人際理論模式應納入關係義務感之構念，才能對憂鬱的人際互動現象有更細緻的瞭解與預測。本論文最後以華人關係特性切入，嘗試討論「台灣與西方之不同樣貌的拒絕」、「關係義務感之調節效果」等論點，另一方面，對本地臨床實務中的憂鬱防治亦作了討論與建議。

AbstractThe present thesis aimed to investigate the responses of relatives and friends toward depressed individuals. Specifically, we postulated that perceived reassurance-seeking (P-RS) and sense of relation obligation (SRO) of the relatives and friends would influence their own negative emotions and rejection. Coyne (1976b) proposed an interpersonal theory of depression which describes the nature of the interaction between depressed people and their significant others. In essence, the interaction between the depressed and their significant others would eventually produce an interpersonal space whereby depressed people's reassurance seeking would breed negative emotions and rejection from their significant others. The theory also asserted that others' negative emotions and rejection would in turn lead to the maintenance and/or deterioration of depressive symptoms in depressed individuals. Joiner and his colleagues further suggested that reassurance- seeking serves as a core element in the interpersonal theory of depression and completed a series of research since 1992 to verify it. Up till now, there has already existed ample of research on the interpersonal theory of depression in the Western literature, while in Taiwan there is considerably insufficient research evidence. Therefore, we designed the present study to examine several predictions elicited from the interpersonal theory of depression. In addition, we believed that emphasis on “relations” between people is unique in Taiwanese culture and expression of rejection may manifest in various forms. It is thus intriguing and important to add two culture-relevant variables, i.e., “sense of relation obligation” and “different expressions of rejection”, into account for the expanded interpersonal theory of depression.

In order to investigate the expanded theory in Taiwan, the present thesis has developed the Perceived Reassurance-Seeking Scale and Multi-phasic Rejection Scale first. Participants of this study consisted of 94 relatives and friends of depressive outpatients. Brief diagnostic interview and the self-report scales were given at individual base. Results of the study indicated that: (1) P-RS and negative emotions were significantly associated with rejection; however, there was a difference in the expression of rejection between the findings in Western culture and that of Taiwanese culture; (2) SRO served as a moderator of the links between the P-RS and rejection; as well as between the P-RS and negative emotions. The findings of the study were discussed in the context of Chinese relational characteristic. Finally, clinical implications for the prevention and treatment of depression were also addressed.

憂鬱人際認知模式之前瞻性探討：憂鬱與臉部情緒之辨識、判斷及認知基模反應
A Prospective Study of Cognitive Interpersonal Model of Depression: Recognition, Judgment, and Schematic Response of Facial Emotions

摘要以憂鬱的人際認知整合取向為背景觀點，本研究的主要目的為探討個體對非語言人際訊息—臉部情緒表情的辨識、判斷及認知基模反應與憂鬱狀態之關連。本研究依循Coyne（1976b）的憂鬱人際論述與Beck（1967）的憂鬱認知理論，推測憂鬱傾向個體能敏感、準確辨識出負向臉部情緒，並對此負向訊息展現負向的認知基模反應傾向；而此負向認知基模反應模式應為穩定認知特質，能與非語言人際壓力共同預測個體的憂鬱狀態。

本研究以54位本地大學／研究所學生為樣本，在資料蒐集的第一階段，進行同時間點憂鬱症狀與臉部情緒之辨識、判斷與認知基模反應等變項表現的測量；七週後，完成其中47位受試者的追蹤評估。

研究結果顯示，相較於非憂鬱個體，憂鬱傾向個體在臉部情緒的辨識與判斷表現上並無顯著錯誤／偏誤傾向；憂鬱傾向個體亦能精準辨識出正、負向臉部情緒訊息。而在同等臉部情緒之辨識與判斷表現的基準上，憂鬱傾向個體在面對負向、甚至所有非正向之臉部情緒訊息時，明顯展現認知概念與情緒向度上的負向認知基模反應傾向。綜合兩階段資料之分析結果顯示，個體對負向臉部情緒表情所展現之認知概念基模反應傾向，呈現獨立於憂鬱狀態變化、並具跨時間穩定的特性，確立其特質地位。進一步探討發現，個體與實驗作業中臉部情緒刺激互動後，其負向認知概念基模反應傾向可預測當下負向情緒的發生。而人際環境中負向臉部情緒訊息之經驗與負向認知概念基模反應傾向的交互作用，能預測個體當時的憂鬱症狀，但無法預測長時間後續的憂鬱變化。

本研究以認知歷程層次與特性觀點、認知特異質—壓力模式統整上述結果，並加入憂鬱的人際行為危險因子，共同討論憂鬱發展、維持或惡化的可能脈絡。最後，提出本研究的貢獻與臨床應用價值，討論可能的研究限制並延伸未來研究方向。

AbstractWithin the framework of the integrated cognitive-interpersonal model of depression, this study aimed to investigate the associations between depression and recognition, judgment, and schematic response of nonverbal interpersonal cues--facial emotions. Based on Coyne’s (1976b) interpersonal model and Beck’s (1967) cognitive theory of depression, the present study proposed that the depressed individuals recognize the negative facial emotions accurately, and respond to negative cues in the negative schematic tendency. This negative schematic set would be a stable cognitive factor; in addition, the interaction of the negative schematic set and nonverbal interpersonal stress would predict depression.

Participants were 54 undergraduate/graduate students. The measurement included depressive symptoms, recognition of facial emotions, judgment of facial emotions, and schematic response to facial emotions. Forty-seven participants were repetitively measured after seven weeks.

The results showed that the accuracy of emotional recognition and preference for emotional judgment did not differ between the dysphoric and nondysphoric individuals. However, the dysphoric individuals tended to respond to non-positive facial emotions with negative cognitive-conceptual and affective schemas. In addition, the cognitive-conceptual schematic response tendency to negative facial emotions was depressed state-independent and did not change over time, and therefore may have trait-like quality. Moreover, after the interacting with the experimental facial stimuli, the increase of individuals’ negative affect could be predicted by their negative cognitive-conceptual schematic response. It was also found that the interaction of this negative schematic response set and the negative facial emotion experiences in the interpersonal environment could predict the individuals’ concurrent depressive symptoms. Contrarily, it could not predict the change of individuals’ depression over time.

In conclusion, the levels and features of cognitive processing as well as the diathesis-stress model were discussed as a synthesis framework. We also combined the results and the interpersonal-behavioral risk factor to discuss the possible context of depression. Finally, we discussed the contributions and clinical applications of the present study, addressed the possible limitations, and provided some suggestions for future researches.

自我感、復原力與創傷後症狀之關聯性研究：以燒傷病人為例
Associations of Sense of Self, Resilience and Posttraumatic Stress Symptoms
among Burn Patients

摘要本研究的目地在於從傳統與正向心理學的角度，以實徵方式探討燒燙傷病患之創傷，並嘗試釐清自我感、復原力與創傷後心理症狀的關聯性。 本研究的參與者為來自南臺灣國軍左營醫院燒傷中心輕中重度單純燒傷病患，共33名，以個別施測方式蒐集資料。

研究中檢驗的假設與分析結果包括下列數項：首先，由於過去的文獻指出外科燒傷所定義之檢傷系統無法解釋燒傷後病患之心理困擾，因此，本研究就此議題進行檢驗，結果發現，外科燒傷嚴重等級與創傷後症狀並無存在顯著相關。

在基本假定理論中，Janoff-Bulman (1985,1992,2006) 認為個體之內在假定為個體組織評估自我，以及評價自我歸屬感與他人連結感之認知表徵。在創傷發生後，創傷衝擊內在對世界與自我的假定而造成心理損害，也就是自我感的損壞。PTSD則被視為個體內在假定受到動搖後所產生之不適應的因應反應。過去的研究結果雖大致符合創傷事件前後個體基本信念產生變動，並與心理症狀呈正相關，但由於研究方法與結果上，仍有諸多岐異，因此本研究以本土發展之自我感量表首度進行理論驗證。結果發現，燒傷前後的自我感變動呈顯著差異，而與創傷後症狀存在正相關，且能顯著預測創傷後症狀。

由正向心理學角度切入，探討有關復原力(resilience)的特性，復原力可被視為一種人的特質，亦是個體經歷逆境中所展現的一種動力過程，強調在顯著的負向情境中個體所表現出的正向適應力。由於過去的復原力研究大都對逆境事件與正向心理適應的界定內容並不一致，測量單一或多重負向事件的方法與正向適應的標準不同，因此本研究嘗試以芬蘭學者Friborg等人 (Friborg, Hjemdal, Rosenvinge, Martinussen, Aslaksen, & Flaten, 2006) 所發展之成人復原力量表 (Resilience Scale for Adults) 做為直接測量復原力的工具。結果顯示，燒傷前後的復原力變動並無顯著差異，燒傷後復原力與創傷後心理症狀存在正相關，且能顯著預測創傷後心理症狀。

無論自我感或是復原力研究之結果，大致符合本研究之假設與過去的文獻理論推論。不過，與研究假設不一致之處，在於復原力的變動量不能預測創傷後心理症狀之能力，而燒傷前的復原力無法預測燒傷後心理症狀，此結果可能導因自研究參與者之特性，而復原力特徵的確在創傷事件發生後才會明顯展現的特質。最後，針對本研究的貢獻與臨床上可能的應用進行討論，並提出研究的限制，與未來可供研究的方向。

AbstractThe main purpose of the present empirical study was to investigate the associations of sense of self, resilience and posttraumatic stress symptoms in burn patients in Taiwan. There were 33 burn patients from Kaoshiung Tsuo-Ying hospital participated in this study.

According to the literature review, there were no direct association between the degree of severity rated by surgeons and postburn psychological problems. This study reexamined this issue and resulted similarly.

The basic assumptive theory, proposed by Janoff-Bulman (1985, 1992, 2006), is about cognitive representation of valuations that organize the self as well as the value of sense of connection and belonging. While the trauma has shattered the assumptions, collapse and breaking-down in self severs as the consequence of damage in the assumptive world. PTSD is then seen as maladjusted response after violation of the assumptions. Some previous studies had shown that the basic assumptions were impacted by the trauma, and also positively correlated to psychological symptoms. However, the measures used in previous studies were inconsistent. This study used indigenous measure of Sense of Self Scale. The results showed that the scores of pre/post burn trauma were significantly different, positively related to posttraumatic stress symptoms, and able to predict the posttraumatic stress symptoms.

Based on the positive psychology, resilience is not only defined as personal characteristic but also dynamic process which leads to positive adjustment to adversity. Due to the inconsistency of definition about adversity and adjustment to single or multiple negative events in previous studies, this study adopted the Resilience Scale for Adults developed by Dr. Friborg（Friborg, Hjemdal, Rosenvinge, Martinussen, Aslaksen, & Flaten, 2006） to measure resilience. The results showed that there were no significant differences between pre- and post-burn resilience; the score of postburn resilience were negatively related to posttraumatic stress symptoms. Results of both sense of self and resilience generally met the hypotheses and were consistent with previous theoretical inference. Nevertheless, it is inconsistent with of the hypotheses about resilience that the total change score between pre- and post-burn Resilience Scale were not able to predict posttraumatic stress symptoms. In addition, preburn resilience was not able to predict posttraumatic stress symptoms.

These results may be because of the diversities of the characteristics of the participants, or the characteristic of resilience may indeed emerge to manifest itself after the traumatic event. Finally, the contributions, limitations, and clinical applications of the present study were addressed in the last chapter. Some possibilities to carry on further researches were also provided.

乳癌婦女的創傷後成長對其安適感之影響：
手術距今時間、診斷期數及人格特質可能的調節角色
The effects of Posttraumatic Growth on Well-Being and Moderating Roles of Time since Surgery, Stage of Disease and Personality Traits in Women with Breast Cancer

摘要創傷後成長 (Posttraumatic growth) 是指個體在和創傷或高壓力事件奮鬥歷程裡，所經驗到的正向改變。然而，在實證研究裡，如此正向的經驗卻與個體主觀安適感缺乏一致的關聯性。故本研究針對一群經歷過乳癌創傷事件的婦女，探討影響創傷後成長與安適感間相關性的可能因素（如: 調節變項），期許能更廣泛地瞭解創傷後成長的現象，特別是在它對個體安適感的影響。

本研究提出四個可能的假設。首先，相較於手術距今時間少於兩年的婦女，當手術距今時間超過二年時，創傷後成長會與安適感呈現較強且較正向的關聯性。第二，對於診斷期數較嚴重的婦女，創傷後成長會與較低安適感（如: 較低生活滿意度和較高的創傷壓力症狀與憂鬱症狀）相關。第三，相較於人格特質較內向的婦女，人格特質較外向的婦女所知覺到的創傷後成長會與較佳安適感相關聯；相較於低神經質的婦女，高神經質的婦女所知覺到創傷後成長會與較差安適感相關。最後，相較於經驗到中等程度創傷後成長的婦女，經驗到低或高程度創傷後成長的婦女將報告較佳的安適感；此意涵著創傷後成長與安適感間的相關性為曲線相關。

此研究為橫斷性研究設計，有效樣本為281位乳癌婦女，選自於乳癌防治基金會或台大醫院。每位受試者均先由外科醫師轉介，在同意參與研究後，請受試者填寫一系列的量表。基於創傷後成長量表之探索性因素分析的結果，在資料分析裡，創傷後成長不僅以總分進行分析，同時也將它解構為兩個子因素（內在成長與人際成長）加以分析。

本研究發現：

1. 手術距今時間僅發現一個調節效果，但卻與預期不符，結果顯現只有在手術距今少於兩年的婦女身上，人際成長與創傷後壓力症狀呈顯著負相關，對於手術距今時間超過兩年的婦女則無顯著相關。

2. 診斷期數與外向性的人格特質並未調節成長與安適感間的關聯性。

3. 神經質的人格特質被證實是成長與安適感間可能的調節變項，不過，仍與假設不符。成長與生活滿意度呈正相關，但如此的關聯性在高神經質的婦女身上更為顯著；此外，成長與憂鬱呈負相關，但僅發生於具有高神經質的婦女身上，對於低或中等神經質的婦女則無顯著相關。

4. 研究結果部份支持成長與安適感呈現曲線相關。相較於知覺到中等程度成長的婦女，知覺到低或高程度成長的婦女傾向報告較少的創傷後壓力症狀與憂鬱症狀。不過，如此的現象並沒有出現在成長與生活滿意度之間的關聯性。

上述結果可能的含意將進一步與過去研究一同討論，並提出未來研究建議。

AbstractPosttraumatic growth (PTG) refers to the experiences of positive changes that result from struggling with a trauma or highly stressful event. However, among empirical studies, such positive experiences were hard to find a consistent association with well-being. Therefore, this research was aimed to investigate possible factors which may moderate or affect the relations of posttraumatic growth (PTG) to well-being in a sample of breast cancer (BC) women in Taiwan. It is hoped that the evidence from this research will further our understanding of the phenomenon of PTG more generally, and the association between PTG and well-being more particularly.

Four hypotheses have been proposed. Firstly, stronger PTG would relate to greater well-being (e.g., higher life satisfaction, and lower PTSD symptoms and depression) when time since surgery is over 2 years. Secondly, PTG would associate with less well-being (e.g., lower life satisfaction, and higher PTSD symptoms and depression) for BC women with more severe disease stage. Thirdly, among extraverted women, PTG would correlate with higher well-being than introverts; in contrast, for women who have high neuroticism, PTG would associate with less well-being than those with low on neuroticism. Finally, women who perceived low or high level of PTG would report greater well-being than those with intermediate level of PTG, which implies a curvilinear association between PTG and well-being.

This is a cross-sectional research design. Total valid sample of BC women with surgery was 281, recruited from the Foundation of Breast Cancer Prevention and Treatment or National Taiwan University Hospital. Each participant was referred from a surgical physician, and after agreeing to take part in this study, they were then given a series of questionnaires. According to the result of exploratory factor analysis for posttraumatic growth scale, PTG was analyzed not only as a total score, but also deconstructed into intra- and inter-personal growth.

Results of the current study indicate that:

1. One moderating effect of time since surgery was found. It revealed that intrapersonal growth was associated with less PTSD symptoms only for women who had surgery within 2 years, but non-significantly for those with time since surgery over 2 years.

2. Stage of disease and Extraversion did not moderate the relationship between growth and well-being.

3. Neuroticism was found to be a possible moderator of the association between PTG and well-being. However, it was contradictory to expectation. Growth related more positively to life satisfaction for women with high level of neuroticism. Furthermore, PTG was correlated with less depression only for high neuroticism women, but non-significantly for those with moderate or low level of neuroticism.

4. Curvilinear relationship between PTG and well-being was partially supported. Women who perceived low or high level of PTG tended to report less PTSD symptoms and depression than those who had intermediate PTG, but such phenomenon was not observed in the association between PTG and life satisfaction.

The implications of these findings are discussed in relation to previous research, and suggestions for future research are further recommended.

Keywords: Posttraumatic growth, Well-being, Breast cancer, Time since surgery, Stage of disease, Extraversion, Neuroticism, Curvilinear relationship.

神經質人格傾向在人際壓力下之因應偏好與情緒因應效能
The Effects of Neuroticism Tendency on Coping Preference and Emotional Coping Efficacy in Interpersonal Conflicts

摘要本研究之目的是瞭解神經質人格傾向在人際衝突下的因應偏好以及衝突後之情緒因應效能。神經質是一個先後被 Eysenck 與 Costa & McCrae 提出來，並被認為是一個代表基本人格向度的人格特質。在神經質量尺得分高的人傾向於容易經驗到負向的情緒，過份的情緒化，對各種刺激的反應過於激烈。神經質傾向被發現與憂鬱、焦慮以及心理疾病有正相關。過去的研究發現，神經質程度不同的人在壓力情境下傾向於使用不同的因應策略。一般認為，神經質傾向可能影響因應偏好，並可能因此影響情緒結果。

為了瞭解神經質的效應，本研究邀請高低兩組神經質傾向受試者（受試者分別為 47 與 46 人）利用每日記錄 (daily rating) 的方式，記錄每天經驗到的負向人際互動的事件，最多三件，並為期一個星期。每日記錄內容包括，衝突事件嚴重度（李克式五點量表）、衝突對象（包括四種關係：父母、手足、朋友或同學、戀人）與衝突種類（包括三種衝突類型：面對面衝突、覺得他人做出傷害自己的事、覺得自己傷害對方）。為了衡鑑每一件衝突事件的因應偏好與情緒狀態，研究者在預備性研究中發展出「負向人際互動因應量表」及「情緒量表」以做為衡鑑工具。經由因素分析，由「負向人際互動因應量表」中找出六個因應策略：面對問題、逃避問題、攻擊性因應、認知調適、尋求社會支持與情緒處理，「情緒量表」中則找出五個情緒類別：憂鬱、焦慮、憤怒、內咎與擔憂等。

本研究發現：

1. 在衝突發生頻率上，高神經質組受試者只在「覺得他人做出傷害自己的事」的頻率顯著高於低神經質組受試者，在「面對面衝突」與「覺得自己傷害對方」的情況則無顯著差異；

2. 在嚴重度評估上，高神經質組受試者也只在「覺得他人做出傷害自己的事」的情況下顯著高於低神經質組受試者，在「面對面衝突」與「覺得自己傷害對方」的情況則無顯著差異；

3. 衝突嚴重程度是神經質傾向影響衝突當時情緒反應強度的中介變項，除了「內咎型情緒」外，其他種類情緒隨衝突愈嚴重而情緒反應愈強；

4. 衝突嚴重程度也是神經質傾向影響衝突因應偏好的中介變項，衝突愈嚴重傾向使用愈多的「認知處理」、「攻擊性因應」與「尋求社會支持」；

5. 衝突對象與衝突種類干擾神經質傾向在「攻擊性因應」上的使用偏好；

6. 神經質傾向在「逃避問題」、「認知處理」、「情緒調適」三種因應策略上則呈現主要效果；

7. 在情緒因應效能上，高神經質受試者使用「攻擊性因應」與「情緒調適」，以及低神經質受試者使用「面對問題」與「認知處理」的方式對情緒反應強度的降低最有效。

本研究最後討論每日記錄的研究方法與憂鬱星期一 (Blue Monday) 的現象對研究結果的可能影響，並對相同議題在未來的研究與在臨床上的應用提出建議。

AbstractThe purpose of the study was to investigate the effect of neuroticism tendency on coping preference and emotional coping efficacy of interpersonal conflicts. Neuroticism is a basic personality dimension posited by Eysenck as well as Costa and McCrae. People who highly on Neuroticism scale is characterized by strong tendency to be subject to negative emotions, over emotionality, and extreme reactions to most stimuli. Neuroticism has been found to be positively related to both concurrent and consequent depression, anxiety, and psychological disorders. Previous research also noted that people with various degrees of neuroticism tend to use different coping strategies while facing stressful situations. It is assumed that Neuroticism tendency may influence the preference of coping and may thus affect emotional outcomes.

In order to understand the effect of neuroticism, the study recruited two groups of subjects with extremely high and low neuroticism (N=47 and 46, respectively), and asked each subject to record at most three conflictive encounters they experienced every day for a week. Daily ratings of severity of each conflict (i.e., 5-point Likert scale), one's role in each interpersonal conflict (i.e., four kinds of relationships: parent/child, sibling, classmate, and romantic intimate), and types of conflicts (i.e., three types of conflicts: face-to-face argument, experiences of being hurt, and experiences of hurting others) were collected. Moreover, in order to assess coping preference and emotional state toward each interpersonal conflict, the Negative Interpersonal Interaction Coping Scale (NIICS) and Emotion Scale (ES) were developed in a pilot study. Through factor analysis, six coping strategies were resolved for the NIICS: facing problems, avoiding problems, cognitive operation, aggressive coping, social support, and emotional modulation, and five emotional clusters for the ES: depression, anger, anxiety, guilt, and mixed anxiety/depression.

The results indicate that

1. In terms of daily frequency of negative interpersonal encounters, subjects of high neuroticism reported significantly more than low neuroticism subjects only on the "experiences of being hurt", but not on the "face-to-face argument" and "experiences of hurting others";

2. With respect to severity rating, subjects of high neuroticism rate significantly more severe than low neuroticism subjects only on the "experiences of being hurt", but not on the "face-to-face argument" and "experiences of hurting others";

3. Severity of conflict serves as a mediator linking neuroticism and coping preference. The more severe the conflict, the more "cognitive operation", "aggressive coping", and "social support" are used;

4. Role in interpersonal conflict and type of conflict both play as a moderator affecting the relationships between neuroticism and aggressive coping;

5. Neuroticism has main effect on three types of coping, namely, "avoiding problem", "cognitive operation", and "emotional modulation", showing subjects of high neuroticism outnumber their counterparts;

6. With respect to coping efficacy, the usage of "aggressive coping" and "emotional modulation" and the usage of "facing problem" and "cognitive operation" has best effects on deduction of negative emotions for subjects of high and low neuroticism respectively.

At last, the study scrutinized the possible effects of daily rating method and Blue Monday phenomenon on the results. Suggestions for future research in this line and clinical implication based on this study were discussed.

華人文化下污名感受、生病行為與情緒對身體化傾向之影響
The Effect of Perceived Stigma, Illness behavior, and Emotion on Somatization in Chinese Culture

摘要本研究以實徵方法探究「身體化(somatization)」傾向的可能成因。不少研究指出華人傾向於透過身體化反應，來表達內心憂鬱與焦慮情緒。經文獻探討發現，華人身體化傾向可能的影響因素有：(1)情緒語彙與情緒表達/壓抑特性；(2)個體對精神問題的社會污名(social stigma)；(3)中西文化傳統中的身心觀點；(4)生病行為之就醫特性。本研究之主要目的乃在：以心理社會因素(psycho-social factor)，解釋台灣民眾至精神科就醫時的身體化現象。

本研究共探究四項重要議題，包括：(1)開場白假說之驗證；(2)表達型及再認型之身、心不適與污名感受間之關聯性探討；(3)情緒字彙、情緒感受/壓抑與身、心不適之關聯性探討；(4)生病行為與身心不適之關聯性探討。

本研究以自行發展及編製的漸進式訪談問卷、污名感受量表與就醫態度量表，以及過去研究探討身心症狀常用的SCL-90-R、身體症狀檢測等量表為工具，並取38名精神科門診輕型精神病患為受試者，得到結果顯示：

1. 約58%患者有先說身體問題再說心理問題的開場白傾向，但開場白傾向未與污名感受達顯著相關，而是與心理諮商需求達顯著負相關。

2. 表達型身體不適與精神醫療污名感受二者間達顯著相關，而再認型身體症狀在去除心理症狀嚴重度之影響後，與污名感受二者間未達顯著相關。顯示污名感受愈強，有愈多身體抱怨，但身體症狀嚴重度不受影響。

3. 情緒字彙極限詞數愈少者，心理症狀嚴重度愈低，而身體訴說字數比例愈高。情緒感受強度較情緒壓抑程度更能預測身心症狀。

4. 一年內身體症狀總數愈多者，愈傾向對自己的問題做超自然歸因。此外，身體訴說比例較高的患者，對精神醫療效果的預期愈高。

本研究進一步探討自行編製使用之污名量表、漸進式病情訪談法及身心訴說內容分析法等研究工具的價值與本研究的限制，並說明未來可以開展的研究方向及臨床應用的意義。

AbstractThis study investigated the possible causes of somatization through empirical method. Some studies pointed out that Chinese people tended to express depression and anxiety through somatization, and the proposed factors accounted for this phenomena including 1) the amounts of emotional vocabularies and characteristics of emotional suppression, 2) the perceived psychiatric stigma, 3) the different conception of body and mind in Chinese and Western cultures, and 4) the characteristics of help-seeking behaviors.

The main purpose of this study was to understand the somatization phenomenon in a psycho-social model, and four hypotheses were examined, i.e., 1) the hypothesis of “prologue theory;” 2) the correlation of the expression mode and recognition mode of illness with perceived stigma, 3) the correlation of emotional vocabularies, emotional suppression, and illness, and 4) the correlation of help-seeking behaviors and illness complaints. This study administered Stepwise Semi-structured Interview Questionnaire, Perceived Psychiatric Stigma Scale, and Help-seeking Attitude Scale that were developed for this study along with Symptom CheckList-90-R and Somatic Symptom Index to 38 neurotic outpatients individually.

The results showed that:

1. 58% patients had the tendency to complain somatic problems before psychological problems. This tendency did not correlate with perceived stigma, but the need of psychological counseling.

2. Somatic complaints correlated with perceived stigma positively. However, after statistically controlling the effect of psychological symptom seriousness, somatic symptoms would not correlate with perceived stigma. It means that the stronger stigma one perceived, the more somatic complaints would be expressed, but neither one affected the actual somatic seriousness.

3. The fewer emotional vocabularies one owned, the less psychological symptoms would be expressed. The extents of perceived emotions predicted symptoms better than the degree emotions suppressed.

4. The more somatic symptoms one experienced in one year, the stronger tendency s/he utilized super-nature attribution. Besides, those who spend much more time in somatic complaints had lower expectation to psychiatric help.

Finally, the instruments developed in this study were discussed and further investigations were suggested.

犯罪青少年與一般青少年之「問題解決歷程」的比較研究
Problem-Solving Model and Coping with Life Stress:
A Comparative Study in Juvenile Delinquency and Normal Adolescents

摘要本研究嘗試以 Frauenknecht 與 Black (1987) 以及 Black 與 Frauenknecht (1990，1994) 所論述之「青少年問題解決模式」為例，分析比較犯罪青少年與一般青少年在實際生活事件中問題解決歷程之特色，藉以瞭解偏差行為發生之可能機制，期能提供相關專業工作者對於青少年偏差行為矯治或犯罪防治上之參考。本研究以46名一般國中生進行前置性研究，以確認研究工具之穩定性與一致性。並且在本研究中選取自台北地方法院少年法庭接受保護管束與桃園少年輔育院接受感化教育的偏差行為青少年共219名，以及正常國、高中在學學生837名，比較不同偏差行為型態之青少年在問題解決歷程中的差異。

研究結果指出下列兩點結論：

1. 有偏差行為之青少年，其整體問題解決效能明顯不佳，進一步分析其問題解決歷程，發現他們與一般正常青少年主要的差異在於「處理問題的自動化歷程」與「解決問題的技巧」兩方面。

2. 透過迴歸分析發現，對不同偏差行為型態之預測最具影響力的因素為「客觀壓力」（學業）、「主觀壓力」（學業／人際）、「處理問題的自動化歷程」(APS) 以及「解決問題的技巧」(CON/EVL) 等六個部分。對憂鬱情緒反應最具預測力的變項主要為「處理問題的自動化歷程」(APS)、「對問題的定向感」(COG/EMO/BEH) 等部分；而對身心症狀嚴重度而言，則為「對問題的定向感」(EMO/BEH) 等兩個部分。

本研究進一步發現，偏差行為青少年解決問題的動機並不匱乏，只是無法有效的運用過去解決問題的相關經驗，且在問題的辨識、決策歷程以及解決問題的評估與監控歷程等解決問題的技巧上較為不足。

最後，研究者提出未來可行的發展方向及本研究結果在臨床應用上的意涵。

AbstractAdolescents at turmoil period have attracted a great deal of public concerns and academic attentions in order to help them to pass through the troubles. Applying Problem-Solving Model, the present study was designed to investigate the possible cognitive mechanism of conduct problems in response to real life events in juvenile delinquencies and normal control. Hence, the results could shed a light on intervention and prevention programs for adolescents with conduct problems.

The Problem-Solving Model, proposed by Black and Frauenknecht (1990 & 1994), consists of three major components: (1) Automatic process, i.e., using past experience to help to resolve the problem; (2) Problem Orientation, i.e., trying to use information to formulate the problem; (3) Problem Solving Skill, i.e., using some skills to put into further evaluations and actions once the problem was mentally formed. Previous western research has found the model applicable to studies of conduct problems.

In order to build up the reliability and stability of the research tools including Social Problem-Solving Inventory for Adolescents (SPSI-A), Life Experience Inventory, Chinese version of Children Depression Inventory (C-CDI) and Psychosomatic Symptom Checklist, this research used 46 junior high school students in a pilot study. In testing the differences between adolescents with and without conduct problems, three samples were used. The group with conduct problems consisted of 219 boys recruited from probation department of Taipei local juvenile court and juvenile reformatory school; the normal control group consisted of 837 boys from a public high school in Taipei.

The results from the pilot study show that research tools have satisfactory psychometric properties so as to ensure reliable utilization in this study. The findings of this study indicated that: (1) there were significant differences of problem-solving efficiency among different groups. In that, the three samples displayed significant differences in Automatic Process and Problem-Solving Skills; (2) according to the regression analysis, the predictors of conduct behaviors included objective academic stress, subjective interpersonal and academics stress, Automatic Process, and Problem-Solving Skills. Besides, Automatic Process and Problem Orientation could predict the depressive mood, and Problem Orientation could predict the severity of somatic symptoms.

To sum up, the present study found that adolescents with conduct problems did not lack motivation to reduce their life stress. Rather, they were in need of efficient ways to utilize past experiences, and had a lack of problem solving skills, such as recognition of problems, decision-making, evaluating and monitoring problem solving process. Based on the findings, the cognitive mechanism of conduct problems, life stresses, problem-solving model, research designs and methodological consideration were discussed. Future research in this line and clinical applications were also suggested.

憂鬱人際模式之探討：尋求再保證的促發、調節與補償機制
Interpersonal Approach to Depression: The Contributory, Moderating, and Countervailing Mechanism of Reassurance-Seeking

摘要

Coyne（1976b）提出的憂鬱人際模式描繪著一個透過與周遭他人的互動影響憂鬱的循環系統。該系統假設憂鬱個體的症狀會引發他人的負向情緒與拒絕，而他人如此的反應則會返身維持／增強個體的憂鬱。在此脈絡下，Joiner及其同僚自1992年開始，透過一系列的研究，進一步提出「尋求再保證」乃憂鬱人際模式的核心；不僅如此，還試圖將尋求再保證由原先「調節」憂鬱的效果的角色，轉為「促發」憂鬱的脆弱因子。整理過往文獻，發現「尋求再保證」概念在論述與研究中有許多混亂的情形，包括變項定義／定位的不清、工具的瑕疵、邏輯的謬誤；因此，本研究經過批判式的分析，提出尋求再保證在不同憂鬱階段中扮演的三種作用：前憂鬱時期的促發作用、憂鬱時期中的補償作用、以及後憂鬱時期的調節作用。

研究分為三個部分，第一部份建立具有信效度的尋求再保證概念測量工具；第二部分則以前瞻式的研究取向，檢驗尋求再保證的促發與調節作用；第三部分以患者與對照組的比較，檢驗尋求再保證的補償作用。

結果發現，

1. 尋求再保證確為憂鬱的脆弱因子，與壓力的搭配之下，透過他人負向反應的中介，可預測憂鬱症狀的發生；

2. 低正向謬誤的憂鬱患者，在尋求再保證下可提高社會支持滿意度，故尋求再保證在個體憂鬱時確有補償作用；

3. 憂鬱的維持，可透過尋求再保證調節初始憂鬱與他人負向反應對後續憂鬱的影響完成。然而，初始憂鬱到後續憂鬱的路徑中，他人負向反應的中介性之不存在，則與原先假設不符。

本論文最後針對不符假設的可能原因及尋求再保證之相關臨床應用考量皆做了討論。

AbstractCoyne (1976b) proposed an interpersonal theory of depression to portrait the nature of the interaction between depressed people and significant others. Such interaction will eventually produces an interpersonal space whereby symptoms of depressed people will breed negative emotion and rejection from others. More important, he believed that these perceived rejections from others would in turn lead to the maintenance and/or deterioration of depressive symptoms in depressive individuals. Joiner and his colleague further postulated that reassurance-seeking serves as a core element in the interpersonal theory of depression. To demonstrate their theory, he and his colleague completed a series of research since 1992 on this topic and their findings mainly showed that reassurance-seeking not only serves to “moderate” the effects of depression, but also plays as a “vulnerability” role of depression. However, in-depth literature review suggests that there still exists a great sense of vagueness about the role and mechanisms of reassurance-seeking in depression arising from unclearly defined variables, inadequate measurements, and the paradoxical hypotheses in past discourses and studies. Therefore, the present thesis introduces three mechanisms of reassurance-seeking on different depression stages through a dialectical analysis process, i.e., contributory effect during the pre-depression period, countervailing effect during the depression period, and the moderating effect during the post-depression period.

Three studies were conducted to examine: a) the reliability and validity of the Chinese version of Reassurance-Seeking Scale; b) the contributory and moderating effect of reassurance-seeking using prospective strategy; and c) the countervailing effect of reassurance-seeking by comparing depressed patients with normal control.

Study 1 reported that the Chinese version of Reassurance-Seeking Scale yielded satisfactory psychometric properties and demonstrated that reassurance-seeking is a reasonably cohesive, replicable, and valid construct. Part 1 of the study 2 prospectively assessed a group of initially symptom-free participants to show that reassurance-seeking predicts future depressive reactions to stress and is mediated by others’ negative responses. The finding from this part is supportive of the vulnerability/ contribution hypothesis of reassurance-seeking. Part 2 of the study 2 prospectively assessed a group of initially depressed participants to investigate whether reassurance-seeking moderates the relation of initial depressive symptoms and negative responses of others with future depressive symptoms. The results are also supportive of the maintainance/ moderation hypothesis of reassurance-seeking. Study 3 demonstrated that reassurance-seeking covaries with social support satisfaction in non-positive illusionlized, depressed people. Our data again provide evidence for the countervailing hypothesis. Specifically, the absence of the mediating effect of “others’ negative responses” of initial depression on future depression is an unexpected result in the study. Some possible explanations for this finding in the context of statistical, conceptual, cultural, and temporal viewpoints will be provided.

Finally, in the last part of the thesis, we discuss some clinical implications of reassurance seeking.

台灣青少年網路成癮之心理病理因子與性別差異：

其與網路使用、壓力、衝動性之關聯性

Etiological Correlates and Gender Differences of Internet Addiction in Taiwanese Adolescents: The Use Pattern, Stress, and Impulsivity

摘要 

目的：隨著台灣網路科技的發達，青少年網路成癮亦成為值得關注之議題。然而，過去研究較少從青少年發展之脈絡與壓力探討網路成癮之成因，更少探討不同類型之壓力對網路成癮之影響；另一方面，過去研究較少解釋男性網路成癮盛行率較高之現象。因此，針對台灣青少年之網路成癮，本研究將先比較不同網路使用活動之預測力，再從壓力-特異質理論檢驗壓力與衝動性之關聯，以期更理解網路成癮之成因與性別差異。本研究假設，網路成癮高危險組之壓力和衝動性均高於低風險組，且壓力調節衝動性與網路成癮傾向之關聯性。進一步地，本研究亦企圖對不同類型之壓力與網路成癮的相關進行探索。最後，針對性別差異，就網路成癮在網路使用、壓力和衝動性之關聯型態加以探討。

方法：研究參與者為台北縣某國中759 名學生(男性387 人，女性372 人，年齡12-15 歲)。參與者及家長均完成同意書填寫，並以匿名方式作答網路使用調查、網路成癮量表(CIAS)、青少年壓力量表，以及貝式衝動性量表(BIS)等自陳式問卷。

結果：透過變異數與階層迴歸分析，本研究發現：(一)青少年男性網路遊戲和情色使用較能預測網路成癮，而女性使用社會互動功能較能預測網路成癮。(二)就壓力類型而言，在青少年男性中，自我與未來擔憂壓力與網路成癮較為相關；相對地，在青少年女性中，浪漫關係壓力與網路成癮較為相關。(三)就客觀學業壓力、主/客觀浪漫關係壓力而言，網路成癮風險與性別有顯著的交互作用。(四)網路成癮高風險組之衝動性高於低風險組，無顯著性別差異。(五)壓力未調節衝動性與網路成癮之間的相關。(六)青少年男性主觀學業壓力與網路成癮傾向之間有顯著的曲線相關。

討論：上述研究結果顯示，台灣青少年男性與女性可能有不同的網路成癮型態與心理病理因子或相關因子。本研究進一步探討青少年網路成癮中，性別因素、壓力與衝動性扮演的可能角色。最後，將論及本研究之限制，並說明未來可行之研究方向與臨床應用。

關鍵字：網路成癮、網路使用、壓力、衝動性、青少年。

Abstract

Objective: While benefitted from highly developed Informational Technology, Internet addiction might be a mental health issue for Taiwanese adolescents and worth research and professional attentions. Nevertheless, environmental and developmental factors associated with Internet addiction in adolescents are less explored. Specifically, little is known about the stress types that may be more related to Internet addiction or the relationship between impulsivity and Internet addiction. Moreover, few studies have attempted to explain higher prevalence of Internet addiction in boys. Therefore, in order to understand the etiological factors and gender differences of Internet addiction in Taiwanese adolescents, the present study attempts to explore the relationship among Internet addiction, stress and impulsivity from a stress-diathesis perspective. It is hypothesized that high-risk adolescents have higher stress and higher impulsivity than low-risk adolescents. In addition, stress is hypothesized to moderate the relationship between impulsivity and Internet addiction tendency. The relationship between different types of stress and Internet addiction is explored. Finally, the gender difference of Internet addiction with respect to Internet use, impulsivity, and stress is explored.

Method: 759 adolescents (387 boys, 372 girls, age range = 12-15 years old) from a local junior high school in Taipei participated in the study. Both participants and their parents completed the informed consent form. Self-report questionnaires assessing Internet usage, adolescents life stress, impulsivity and Internet addiction tendency were completed by participants anonymously.

Results: Using analyses of variance and hierarchical multiple regressions, the present

study found: 1) While online gaming and using cyberporn were more predictive of Internet addiction for boys, using social interaction device was more predictive for girls. 2) Stress of self and future concern was more related to Internet addiction for boys, while romantic relationship stress was more related to Internet addiction for girls. 3) Significant gender × group (low/high risk) interaction was found for objective academic stress and subjective/objective romantic relationship stress. 4) Both high-risk boys and girls showed significant higher impulsivity than low-risk group. 5) Stress did not moderate the relationship between impulsivity and Internet addiction. 6) Subjective academic stress had a curvilinear relationship with Internet addiction, whereas such relationship existed only for boys but not for girls.

Discussion: In Taiwanese adolescents, boys and girls may have different etiological

patterns of Internet addiction. The roles of Internet use, types of stress, and impulsivity playing in the etiological links to Internet addiction are addressed. The limitation of study as well as clinical implication and future direction are discussed.

憂鬱症患者之非語言情緒辨識表現初探
A Clinical Investigation of the Nonverbal Emotional Recognition in Depressed Patients

摘要本研究目的在於以實徵研究方式探討憂鬱與非語言情緒辨識之關聯性，檢驗處於憂鬱發作之憂鬱患者在臉部表情、聲音語調與雙重管道上之非語言情緒辨識表現。在文獻回顧中，皆發現憂鬱個體在非語言情緒辨識上有異於正常個體之表現，但其差異內容因研究對象、研究工具、及研究設計之差異，而在不同研究之間存在歧異性。是故，本研究改善研究設計與方法，並以本土性憂鬱患者檢驗憂鬱症與非語言情緒辨識之關聯性，期望能進一步釐清其可能的關連性。

本研究假設：相較於正常個體，憂鬱個體在單一管道（包括臉部表情和聲音語調）的非語言情緒辨識上，傾向呈現負向偏誤與正向偏誤；在雙重管道（包括一致情緒和不一致情緒）的非語言情緒辨識上，則傾向呈現負向辨識與敏感辨識。本研究的樣本為30名憂鬱患者與 39名社區居民。研究過程中先以結構式訪談問卷篩選出適宜受試者，其後以自填式量表衡鑑其憂鬱情緒狀態，以電腦化非語言情緒辨識工具測量受試者在臉部表情、聲音語調、及雙重管道之非語言情緒辨識能力，並以短版魏氏成人智力測驗第三版評估智商。

研究結果顯示，在排除智商的影響後，憂鬱患者相較於正常者在單一管道的非語言情緒辨識上有負向偏誤，對悲傷負向情緒的辨識較為正確，以及易將其他非語言情緒錯誤辨識成悲傷與憤怒等負向情緒，但在高興正向情緒上則和正常者無顯著差異。換言之，本研究結果支持憂鬱患者具有情緒一致傾向的假設，憂鬱患者傾向增強與情緒一致之負向情緒刺激的處理，對負向情緒具有高敏感辨識度。在雙重管道上，憂鬱患者相較於正常者，對一致情緒訊息與不一致情緒訊息的正確辨識度與正常者無顯著差異；然而，在表情語音雙重管道一致的高興或悲傷情緒之辨識正確率上優於正常組，在一致的憤怒或恐懼情緒之辨識正確率則低於正常組；此外，憂鬱患者在表情語音雙重管道一致的情緒辨識錯誤特性上，較容易出現高興正向情緒之辨識傾向，而在表情語音雙重管道不一致情緒的辨識錯誤特性，則容易出現恐懼負向情緒之辨識傾向。

最後，本研究以認知理論、人際理論、及特質壓力模式等角度討論憂鬱可能之維持機制，並且討論研究結果在臨床應用之意涵，以及可供未來後續研究參考之方向。

AbstractThis study aimed to empirically investigate the association between depression and nonverbal emotional recognition, and examine the possible deficits of nonverbal emotion recognition abilities in facial, paralanguage, and dual-channel expression in depressed patients. Literature review shows that there exists difference of nonverbal emotion recognition between depressed patients and normal controls. However, the results are inconclusive yet. Such inconsistent results probably arise from diverse samples, various tools, and different research designs of existent studies.

Thus, through bettering the research design and assessment tools, the present study used clinically depressed patients and normal controls to examine two hypotheses: a) clinically depressed patients might display negative and positive biases in various emotion recognition of both facial and paralanguage expression; b) clinically depressed patients might show negative and sensitive tendency in dual-channel nonverbal emotion recognition task. The subjects consisted of 30 clinically depressed patients and 39 normal controls. The study used structured diagnostic interview for screening purpose, self-report symptom scale for assessing depressive tendency, computerized Diagnostic Analysis of Nonverbal Accuracy 2-Taiwan version (DANVA2-TW) for assessing ability of nonverbal emotional recognition, and short form of WAIS-III for assessing IQ.

Results revealed that after controlling for IQ, there existed negative bias in nonverbal emotional recognition of facial and paralanguage expression in depressed patients. Specifically, depressed patients recognized sad emotions more accurately. They tended to misperceive nonverbal emotions as negative ones including sad and angry, whereas there was no difference in positive emotions. The results suggested that depressed individuals tended to correctly recognize negative emotions and negatively perceive other’s emotions in a mood-congruent manner. Besides, results revealed that, in dual-channel nonverbal emotion recognition, there was no difference in accuracy between depressed patients and normal controls. But, depressed patients tended to be more accurate in recognizing congruent sad and happy emotions and less accurate in congruent angry and fear emotions. In terms of their error pattern, depressed patients tended to misperceive congruent emotions as happy ones, and misperceive incongruent emotions as fear ones.

Finally, some possible explanations about maintenance of depression were discussed from viewpoints of cognitive, interpersonal, and diathesis-stress models. Clinical applications as well as future research directions were also addressed.

過度概化記憶與創傷後壓力症狀之關聯性探討：以九二一地震為例
The Research of Association between Overgeneral Memory and Posttraumatic Stress Symptoms with the adult survivors of the 921 Earthquake

摘要本研究的目的在於以實徵研究的方式，針對九二一震災之創傷並以認知的角度，嘗試釐清過度概化記憶與創傷後壓力症狀的關聯性。創傷後壓力症狀以侵入與逃避兩症狀為主，症狀的展現反映著受創者對創傷訊息的處理過程與狀態。在臨床治療與理論上，個案對創傷的具體化陳述為治療必經之途，且有助於創傷後壓力症狀的緩解。實徵研究上也發現，創傷個體之自傳式記憶有過度概化的現象。綜上推之，自傳式記憶之過度概化與兩大創傷後壓力症狀之間存有相關性。然而，針對此議題之實徵研究數量有限，且研究結果尚無定論。再者，大多研究所測之自傳式記憶為情緒事件，而且少有區分創傷事件的類型，同時以AMT典範測量自傳式記憶的過度概化也缺乏生態效度。是故，本研究針對特定類型的創傷事件，選用與發展合宜之研究工具與程序，對過度概化記憶與創傷後壓力症狀之關聯性進行探索性實徵研究。

本研究樣本為經歷九二一地震之南投縣埔里鎮成年居民192位，針對九二一地震之經歷進行回憶記錄，用改良之自傳式記憶內容概化測量指標進行自傳式記憶概化的客觀分析。再者，進行事件衝擊量表中文版之心理計量檢驗，並以事件衝擊量表中文版測量個體對九二一地震的逃避與侵入症狀的發生頻率。另外，考量創傷後壓力症候群與憂鬱症之高共病與相似之認知特性，用SCL-90-R之憂鬱分量表測得憂鬱症狀且予以排除。研究結果顯示，整體過度概化記憶與兩大創傷後壓力症狀之關聯不顯著。但是，記憶過度概化測量中的部分指標，像是非單一事件、無事件後果、持續時間超過一天、無事件發生地點、地點重複等指標，與侵入症狀呈顯著負向關聯；而逃避症狀與非具體描述、持續時間超過一天、無事件發生地點、地點重複、無特定人物等指標有顯著正向關聯。上述結果的顯著性，不受憂鬱症狀的影響。最後，本研究以認知角度切入討論研究結果之意涵，並且報告本研究之限制，及提出未來可供後續研究的方向。

AbstractThis study aims to investigate the possible association between overgeneral memory and posttraumatic stress symptoms with earthquake survivors in Taiwan. Intrusion and avoidance, as the main posttraumatic stress symptoms, reflect the cognitive processing state of traumatic information in survivors. Many empirical studies have found overgenerality of autobiographical memory in survivors. Both clinical therapists and theorist think that remission of posttraumatic stress symptoms may accompany more specific description of traumatic event in traumatized clients. Therefore, we infer that there may be plausible association between overgeneral memory and posttraumatic stress symptoms in Earthquake survivors. However, there exists very few empirical studies and most have resulted inconsistently. In addition, most measurements of autobiographical memory fail to specify trauma categories, and the most broadly used AMT paradigm lacks sufficient ecological validity in measuring overgeneral memory. Thus, we conduct the thesis with more sufficient and valid tool and procedure.

The participants of this study are 192 adult survivors residing near the epicenter who were recruited three years post the earthquake. The Impact of Events Scale- Chinese Version (IES-C) was used to assess intrusive and avoidance symptoms, the depression subscale of SCL-90-R was used for depressive symptoms, and modified context-based autobiographical memory task was used to assess the memory about 921 earthquake. Results revealed that various indexes of overgeneral memory were found to have various associations with different elements of posttraumatic stress symptoms. More specifically, intrusive symptom correlated negatively with the indexes of not-single event, no-outcome, duration more than one day, no-place, and repeated place. Avoidance symptom correlated positively with the indexes of not-concrete description, duration more than one day, no-place, repeated place, and not-specific person. Moreover, although there is evident comorbidity between PTSD and depression, the above-mentioned significant results hold after controlling for depression. Finally, discussion on the limitation of the present study as well as possible follow-up research is provided.