社交焦慮者對情緒表情的注意力偏誤：
檢驗正負向評價訊息的效果

研究背景與目的：社交焦慮症的認知行為理論假設社交焦慮者處於社交情境下，訊息處理歷程中容易產生認知偏誤，會將注意力過度地投注在可能隱含負向評價的訊息上，注意力偏誤與害怕評價被證實是發展與維持社交焦慮的重要因子。在過去研究社交焦慮者的注意力偏誤會加入社交情境的操弄來誘發社交焦慮，但對於威脅刺激的注意力偏誤的結果並不一致，為了觀察注意力偏誤的歷程變化也加入不同時距的測量。近年來也有許多研究探討害怕評價對社交焦慮者的影響，也證實社交焦慮者會害怕正向與負向評價，然而較少探討到正負向社交情境對於注意力偏誤的影響。本研究目的在於檢驗社交焦慮者在接收到正向與負向社會評價的情境中，對於不同情緒價性的臉部表情是否會產生注意力偏誤，以及注意力偏誤型態為何，以釐清社交焦慮者的害怕正負向評價特性對其注意力處理的影響。研究方法：研究參與者總共83人非臨床樣本之大學生，透過李氏社交焦慮量表自陳版(Liebowitz Social Anxiety Scale-Self Report, LSAS-SR)進行開篩選高社交焦慮組42人與低社交焦慮組41人，再將其隨機分派到正向情境的高社交焦慮組有20人；負向情境的高社交焦慮組有22人；正向情境的低社交焦慮組有20人；負向情境的低社交焦慮組有21人。參與這被告知進行即席演講的作業，演講結束後給予正向或負向回饋，並進行點偵測的注意力偏誤作業，最後再填寫量表。參與者在進行實驗前、告知演講後、演講結束後、接收評價後與實驗後五個時間點進行焦慮、害怕、生氣、難過與快樂等情緒評估。研究結果：高社交焦慮組的注意力較低社交焦慮組隨著時間的變化對於情緒臉孔更有注意力朝向、注意力脫離困難與注意力逃避的現象。注意力偏誤指標方面，在接受正向情境的高社交焦慮組比接受負向情境的高社交焦慮組，對於正向臉孔呈現1250毫秒會有注意力逃避的現象。注意力投入指標方面，在正向情境中，高社交焦慮組比低社交焦慮組的注意力投入指標分數低，且分數為負值，表示在接受正向情境的高社交焦慮組，對於刺激呈現1250ms的正向臉孔有注意力投入較慢的現象；在高社交焦慮組中，正向情境組的注意力投入分數顯著低於負向情境組，顯示接受正向情境的高社交焦慮組，對於刺激呈現1250毫秒的正向臉孔會有注意力投入較慢的現象。注意力抽離指標方面，在正向情境中刺激呈現500 ms時，參與者對正向臉孔相較於威脅臉孔更有注意力脫離困難的現象。結論：在正向情境下，當正向刺激呈現1250毫秒時，高社交焦慮組相較於低社交焦慮組明顯的對於正向刺激更有注意力逃避的現象；且相較於負向刺激，對正向刺激有注意力投入延宕的現象。

關鍵字

社交焦慮；害怕負向評價；害怕正向評價；注意力偏誤

並列摘要

Background and Aims: According to cognitive behavioral theory of social anxiety disorder posits that, in social contexts, sufferers tend to generate information processing bias, paying excessive attention to potentially negative information. Fear of evaluation and attentional bias play a pivotal role in the development and maintenance of social anxiety disorder. In the past, studies on social anxiety related attentional bias have manipulated a given social situation to trigger anxiety in research subjects. The study results on attentional bias connected to threatening stimuli, however, are inconsistent. Different temporal measurements have also been devised to observe the full gamut of changes in attentional bias. In recent years, a significant number of studies have investigated the influence of a fear of evaluation on sufferers of social anxiety disorder, proving that sufferers fear both positive and negative evaluation. A relatively small number of studies, however, have investigated the influence of positive and negative social situations on attentional bias. The aim of the present study is to test whether socially anxious subjects will generate an attentional bias in relation to the affective valences of different facial expressions after receiving a positive or negative social evaluation. We also seek to determine the type of attentional bias generated in order to shed light on the influence of both positive and negative evaluation on the attentional processing of subjects. Methods: A total of 83 university students were enlisted as nonclinical samples. After conducting the Liebowitz Social Anxiety Scale-Self Report (LSAS-SR), 42 subjects were assigned to a high socially anxious (HSA) group and 41 to a low socially anxious (LSA) group. 20 subjects from the former group were randomly sent to the “Positive Social Situation,” and 22 to the “Negative Social Situation.” 20 subjects from the latter group were sent to the “Positive Social Situation,” and 21 to the “Negative Social Situation.” Subjects were told that they were expected to give a spontaneous speech, following which they would receive positive or negative feedback. They were also expected to comply with spot checks for attentional bias and, finally, complete a scale. Emotional assessments of the subjects’ anxiety, fear, anger and sadness were conducted at five points: before the experiment, after being informed about the speech, after giving the speech, after receiving evaluation and after finishing the experiment. Results: Over the course of time, subjects from the HSA group showed greater degrees of directed attention, attentional disengagement, and attentional avoidance than subjects from the LSA group. In regard to the Attention Bias Index, subjects from the HSA group displayed clearer signs of attentional avoidance when shown a positive face for 1250 milliseconds than subjects from the LSA group. In regard to the Attention Engagement Index, subjects from the HSA group registered lower attention investment scores (and even negative values) in positive social situations than subjects from the LSA group. This shows that in positive social situations, subjects from the HSA group displayed slower attentional engagement when exposed to the stimulus of a positive face for 1250 milliseconds. Among subjects from the HSA group, the attention engagement score in the positive social situation was markedly lower than among subjects from the same group in the negative social situation, showing that subjects from the HSA group displayed slower attentional engagement in response to the stimulus of a positive face for 1250 milliseconds. In regard to the Attentional Disengagement Index, subjects exposed to a stimulus for 500 milliseconds displayed a greater degree of attentional disengagement to positive faces than threatening faces. Conclusion: In positive social situations, subjects from the HSA group displayed a markedly greater degree of attentional avoidance when exposed to a positive stimulus for 1250 milliseconds than subjects from the LSA group; moreover, these same subjects showed a greater degree of attentional engagement delay in response to a positive stimulus than to a negative stimulus.