情緒脈絡線索與表情的一致性在社會焦慮者表情辨識所扮演的角色

Abstract

研究目的與方法：本研究旨在同時探討：情緒操弄、文字情緒線索與表情的一致性，對高、低社會焦慮者表情辨識能力的影響。本研究為2（社會焦慮組別：高社會焦慮、低社會焦慮）× 2（情緒操弄：虛擬演講組、放鬆組）× 5（表情種類：中性、快樂、生氣、厭惡、輕視）× 2（文字情緒線索的一致性：一致、不一致）的四因子混合設計。其中，表情種類與文字情緒線索的一致性為受試者內變項，社會焦慮組別與情緒操弄組別為受試者間變項。而依變項則為表情辨識的正確率、錯誤率與反應時間。本研究使用害怕負向評價量表（FNE），將受試者分為高社會焦慮傾向組42人、低社會焦慮傾向組41人，使用生氣、輕視、厭惡、快樂、中性此五種表情，將文字情緒線索和表情進行配對，並先後呈現，產生文字情緒線索和表情一致以及不一致的嘗試，並藉由虛擬演講或放鬆指導語，操弄情境焦慮的狀態。結果與結論：結果發現，雖然受試者在不一致的嘗試中，表情辨識的正確率較一致嘗試差，但是低社會焦慮組在一致嘗試與不一致嘗試兩者間，其正確率差異較大，顯示低社會焦慮組較易受先前呈現的文字情緒線索而影響判斷，而高社會焦慮組較能針對表情本身作反應，不易受文字情緒線索影響。此外，情緒操弄對正確率造成的影響不顯著，但於反應時間上，在不一致的嘗試中，虛擬演講組顯著慢於放鬆組。另外，在錯誤率當中，高、低社會焦慮組在輕視表情上有不同的錯認傾向，高社會焦慮組傾向將其錯認為中性，其次為厭惡表情，而低社會焦慮組則傾向將其錯認為厭惡以及中性表情，兩者機率差不多。依據高社會焦慮者較不易受情緒線索影響表情判斷的結果，若推論至社交情境中，其可能有傾向於關注於他人負向評價，而未能觀察情境中其他不一致線索的現象。此外，高社會焦慮者傾向將輕視表情辨識為中性，亦可能顯示其對不具威脅性之社交回應的期待。因此，若將本研究之發現，運用在社會焦慮者的治療，可把重點放在高社會焦慮者練習以較全面性的方式蒐集社交訊息，以及對於他人評價的減敏感。

Purpose and method: The purpose of this study was to investigate the interplay of situation manipulation and the congruency of emotion contextual cues and facial expression on recognition of facial expressions in different social anxiety groups. A 2 × 2 × 5 × 2 mixed design with social anxiety group (high vs. low) and situation manipulation (bogus-speech vs. relax) serving as between participants factors, and facial expression category (angry, contempt, disgust, happy, and neutral face) and congruency (congruent vs. incongruent) serving as within participants factors. The dependent variables were accuracy, error rate, and reaction time. A total of 83 participants were divided into high social anxious group (n = 42) and low social anxious group (n = 41) according to their scores on the Fear of Negative Evaluation Scale. To explore the influence of facial expressions, emotional context, and their interplay on recognition accuracy, five facial expressions along with their five correspondent emotional words were used to form congruent and incongruent pairs, respectively. Results and conclusions: There was a significant influence of emotional context; both groups made more mistakes in incongruent pairs than in congruent ones. The interaction of social anxiety groups and congruency implied that low social anxious participants were more prone to be influenced by emotional context than high social anxious participants. While facial expressions and emotional context both played a role in accuracy of facial expression recognition, high social anxious participants were less prone to be influenced by emotional context, which resulted in more precise in recognizing facial expressions. Besides, while situation manipulations did not made the difference significantly on accuracy of facial recognition, the reaction time for bogus-speech group was significantly longer than that in relaxed group under incongruent pairs. As to error rate, the misrecognition prone of contempt was different in the two social anxiety groups. High social anxious participants were more prone to recognize contempt as neutral than disgust, whereas low social anxious participants were prone to recognize contempt either as disgust or as neutral. According to our study, high social anxious participants were less prone to be influenced by emotion context, which indicated that in social conditions, high social anxious individuals might focus on negative social context and ignore other social contexts as well. Besides, high social anxious participants were more prone to recognize contempt as neutral, which might indicate the expectancy of non-threatening social reply. Taken together, when doing psychotherapy for social anxious individuals, the importance of enhancing their breadth of attention in collecting social information and desensitization of negative evaluation are discussed.