慢性病患及高低特質焦慮者之注意偏誤及其焦慮反應間的關係

The Relationship between Attentional Bias and Anxiety in Chronic Disease Patients and High- and Low-Trait Anxiety

黃芸新(Yun-Hsin Huang)

指導教授 ： 吳英璋

國立臺灣大學/理學院/心理學研究所/碩士(2010年)

https://doi.org/10.6342/NTU.2010.02982

摘要

罹患高血壓、高血糖或高血脂等疾病的慢性病患者，針對「健康」之過度焦慮會影響其疾病症狀與相關行為，因而妨礙疾病控制。過去許多研究發現，具有臨床焦慮疾患、高特質性焦慮或處於高焦慮狀態之個人，對於威脅刺激皆有注意偏誤。威脅刺激之注意偏誤可進一步區分為「過度警覺」及╱或「脫離困難」，此兩類偏誤來源皆可能造成焦慮之發生。根據過去實徵研究結果，以及已被提出的兩個用以解釋注意偏誤之機制，本研究提出一整合性的模式，嘗試解釋注意偏誤之來源。本研究主張工作目標與具威脅性的分心刺激皆會形成內在表徵，並接受不同內在系統之輸入：對工作目標之意識努力會增加其表徵強度，而自動化威脅評價系統及意識化威脅評價系統之輸入則會強化威脅刺激之表徵。兩個內在表徵會互相競爭，若威脅刺激較佔優勢，則個體即會注意威脅刺激。本研究主張，「過度警覺」即反映威脅評價系統之閾值過低，而「脫離困難」則反映對工作目標投注之意識努力不足，兩者皆會影響焦慮情緒之發生。本研究之目的即在於以上述指標對於焦慮反應（包含自陳式量表與生理心理指標）之預測性，檢驗本研究提出之假設模式是否被支持，並嘗試探索逃避注意威脅刺激時，情緒中的不同面向於注意偏誤與焦慮間所扮演之角色。
本研究於兩個地區性基層家庭醫學科診所，募集到30位慢性病患者（16位男性，平均年齡53.23歲）以及27位非慢性病患之參與者（5位男性，平均年齡40.11歲）。本研究使用多個焦慮指標，包括兩個自陳式量表（特質焦慮量表與焦慮敏感度指標）、立即性皮膚電反應、皮膚電反應恢復困難程度、立即性額頭肌肉電反應、額頭肌肉電反應恢復困難程度等，其中焦慮生理指標之測量與空間線索導引作業同時進行。「空間線索導引作業」以「威脅詞」、「模糊詞」與「中性詞」作為實驗刺激。威脅詞與模糊詞皆與中性詞配對，一半配對呈現100毫秒，另一半則呈現1500毫秒，以分別反映自動化歷程與意識化歷程。根據反應時間可計算三類指標：（1）100毫秒警覺指標，反映自動化威脅評價系統之輸出；（2）1500毫秒警覺指標，反映意識化威脅評價系統（含自動化威脅評價系統）之輸出；（3）1500毫秒警覺指標，反映對工作目標付出的意識努力。為了驗證假設模式，本研究以前述三類注意偏誤指標預測焦慮指標，並將受試者分為「慢性病組」與「非慢性病組」，以及「高特質焦慮組」與「低特質焦慮組」，進行多個複迴歸分析。
結果發現：（1）假設模式中的意識化威脅評價系統與意識努力不足皆能預測多項焦慮生理反應指標，尤其是皮膚電反應相關指標；（2）自動化威脅評價系統可預測慢性病患之特質性焦慮；（3）慢性病患者與低特質焦慮者對模糊刺激之逃避偏誤，可預測較大的立即性肌肉電反應。本研究主張假設模式中的兩個意識化歷程與焦慮生理反應之發生有關，但其與較長期的特質性焦慮之間的關連可能受到因應之影響，低特質焦慮者即可能利用「逃避中低度威脅刺激」的方式進行因應；而關於快速呈現之刺激在中高齡慢性病患者所引發的焦慮生理反應，則需進一步探討。此外，本研究進一步認為，在低特質焦慮者，「皮膚電反應」可能與「傾向注意威脅刺激」有關，而「肌肉電反應」則與「傾向逃避威脅刺激 」有關，並據此主張情緒中的不同面向在注意偏誤歷程中可能扮演不同的角色。

關鍵字

慢性病 ； 注意偏誤 ； 焦慮 ； 情緒生理反應 ； 空間線索導引作業 ； 自動化威脅評價 ； 意識化威脅評價

並列摘要

Being over anxious about health can be harmful for patients with hypertension, diabetes mellitus, or hyperlipidemia. Attentional bias toward threat stimuli, especially mild threatening ones, has been found important in generating anxiety. However some researches have suggested that high trait-anxiety participants tend to consciously avoid high threatening stimuli. Bias toward threat stimuli could be further divided into over-engagement and/or disengagement difficulty. The present study proposed integrating model of two previous models. This model suggested that internal representation of a threat stimulus receives input from both preattentive and guided/conscious threat evaluation system. In contrast, voluntary effort strengthens target representation. The representations compete with each other for attention. It was hypothesized that over-engagement reflects low threshold of threat evaluation system, and disengagement difficulty reflects insufficient effort. Accordingly, these two indexes both have contribution in generating anxiety. The objective of this study is to investigate if the proposed integrative model is supported by examining the predictiveness of those indexes on anxiety response (including self-report and psychophysiological indexes).
Thirty subjects recruited from two local family medicine clinics were grouped based on diagnosis of chronic diseases (the three diseases described above). Sixteen of them are male. Mean age is 53.23. Another 27 subjects without chronic disease were recruited and served as control group. Five of them were male. Mean age is 40.11. Each subject completed two self-report anxiety questionnaires: Spielberger’s State-Trait Anxiety Inventory-Trait version, and Anxiety Sensitivity Index. Then the spatial cueing task would be done while his/her electrodermal activity(EDA) and electromyography(EMG) was measured, which reflect arousal level of autonomic nervous system. The spatial cueing task was designed with three kinds of materials: threat, ambiguous, and neutral words. Words presented in pair for 100 or 1500 millisecond(ms), reflecting the preattentive and conscious system respectively. Therefore three indexes are concerned: 1) 100ms engagement, as indicator of preattentive threat evaluation system (PTES); 2) 1500ms engagement, reflecting function of guided, including prior preattentive, threat evaluation system (GTES); 3) 1500 ms disengagement, indicating the input from guided effort toward target. Participants are divided into 1) chronic disease vs. non-chronic disease; 2) high trait anxiety vs. low trait anxiety. Multiple regression analyses were conducted with three attentional bias indexes as predictor and different anxiety responses indexes as dependent variables in each group.
The results implicated: 1) GTES and insufficient guided effort are able to predict multiple anxious psychophysiological indexes, especially EDA-related ones; 2) Trait anxiety of chronic disease patients could be predicted by PTES; 3) Tendency to avoid ambiguous stimuli in chronic disease and low trait anxiety groups predicts stronger immediate EMG responses. The results suggested that two conscious processes in the hypothesized model are related to anxiously psychophysiological responses, but the relationship to trait anxiety might be influenced by coping. Furthermore, the result implicated that EDA responses are related to attentional bias, and EMG is related to avoidant bias in chronic disease and low trait anxiety groups, thus different components in emotion might be related to different tendency in attention process.

並列關鍵字

Chronic disease ； Attentional bias ； Anxiety ； Emotion ； Psychophysiology ； Spatial cueing task ； Preattentive threat evaluation ； Guided threat evaluation

參考文獻

陳明豐（1998）。「冠狀動脈與心臟病」。台北：健康世界雜誌。

鄭逸如、吳治勳、黃芸新（2008）。慢性病患者之死亡威脅、慢性病威脅與憂鬱：個人建構理論的觀點。「中華心理衛生學刊」，21，51-73。

鄧閔鴻（2010）。「廣泛性焦慮疾患的心理病理：病態憂慮、注意偏誤、與憂慮的後設認知」。台北：國立台灣大學心理學研究所博士論文。

顏孝羽（2006）。「內隱記憶偏誤於恐慌症病理之角色」。台北：國立台灣大學心理學研究所碩士論文。

American Psychiatric Association, (1994). Diagnostic and Statistic Manual of Mental Disorders (4th ed.). Washington, DC: Author.