台灣地區華人情緒與相關心理生理資料庫－－大學生基本情緒臉部表情資料庫

Taiwan Corpora of Chinese Emotions and Relevant Psychophysiological Data–A College-Student Database of Facial Expression for Basic Emotions

本研究之主要目的係建立一個以台灣大學生為主的華人基本情緒臉部表情資料庫，共分成兩個部分進行。在第一部分，我們藉由標準化的方式廣泛地蒐集48名大學生模特兒臉部表情的影像，男女各24名；接著藉由主觀的行為評量建立各類表情的常模，此一常模包括強度清晰的情緒表情及強度較弱的表情，共計170位受試者參與評量。評量之結果則依據強度類別及多數類別加以分類臉部表情，並針對個各表情評量之強度類別結果計算其平均熵值（average entropy）。本研究的第二部分則是蒐集基本情緒的自然臉部表情刺激材料，共蒐集65位大學生（含研究生）模特兒，其中男性33位，女性32位。實驗程序中引發模特兒的基本情緒反應，並攝製其自然臉部表情。然後對所蒐集之自然臉部表情進行評量，共計84位受試者參與評量。評量之結果亦依據強度類別及多數類別加以分類，並亦針對個各表情評量之強度類別結果計算其平均熵值。本研究所建立之基本情緒臉部表情資料庫，將可提供其他研究者探討臉孔表情及相關情緒議題之用。

The main goal of this study was to establish a database of facial expression for basic emotions among Chinese using college students in Taiwan. The study consists of two parts. In Part 1, we employed a standard procedure and collected a wide range of facial images representing expressions of basic emotions from 48 college students who served as models, half males and half females. Following that, we established a norm for each kind of basic emotions, consisting of relatively strong and relatively weak expressions, based on subjective behavioral ratings from another group of 170 participants. The results of rating were used to classify each facial expression into different categories in accordance with criteria of intensity and plurality, and the mean value of entropy also was calculated for each category of facial expressions according to intensity rating. In Part 2, we collected facial expressions of basic emotions from 65 college students (both graduate and undergraduate), 33 males and 32 females, who served as models under naturalistic settings. Experimental procedure was used to induce emotional reactions from the models, from which images of facial expressions were extracted. The naturalistic facial expressions were later rated by another group of 84 participants. As in Part 1, the results of rating were used to categorize facial expressions into basic emotions and to calculate their respective entropy values of each category. The database of facial expressions for basic emotions established in the present study can provide a valuable research tool for investigators who are interested in studying facial expressions and issues related to emotional processing.