人類表情豐富超乎你想象 共有21種表情

When you're feeling one of those combination emotions, like sadly angry or happily disgusted, it shows on your face in a unique way, a new study shows.

In fact, there may be at least three times more recognizable human facial expressions than previously thought, the researchers said.

In a new study, researchers defined 21 facial expressions that we use to convey our emotions, and found that a computer model could tell them apart with a high degree of accuracy by looking for small changes in facial muscles.

Among these facial expressions were the six, long-recognized basic emotions — happy, sad, fearful, angry, surprised and disgusted — but also 15 others, which were a combination of these basic feelings. For example, a person may show that they feel happily surprised, or angrily surprised, said the researchers, calling such feelings compound emotions.

This extended library of facial expressions of emotions may be useful in studying human brain and social communication, as well as in the design of computer systems that can communicate with humans, researchers said in their study published today (March 31) in the journal Proceedings of the National Academy of Sciences.

"We've gone beyond facial expressions for simple emotions like 'happy' or 'sad.' We found a strong consistency in how people move their facial muscles to express 21 categories of emotions," said study researcher Aleix Martinez, a cognitive scientist and associate professor of electrical and computer engineering at The Ohio State University. "That tells us that these 21 emotions are expressed in the same way by nearly everyone, at least in our culture."

Historically, scientists and philosophers have focused their studies on six basic emotions, but deciphering a person's brain functioning with only six categories is like painting a portrait with only primary colors, Martinez said in a statement.

In the study, the researchers took about 5,000 photographs of 230 college students who were asked to make faces in response to verbal cues such as, "You just got some great, unexpected news" (happily surprised) or "You smell a bad odor" (disgusted).

To determine which expressions were unique enough to be consistently distinguished from others, the researchers analyzed the photographs with a computer program called the Facial Action Coding System (FACS). Popularized by psychologists in the 1970s, FACS breaks a facial expression down into its elements, such as raised eyebrows or wrinkled nose, and analyzes the underlying muscle movements that are used to create that expression.

The FACS analysis revealed that the 21 expressions used a unique combination of muscles that was different from all other expressions. A computational model of face perception identified the six basic expressions with 96.9 percent accuracy, and the 15 compound expressions with 76.9 percent accuracy.

Some expressions were more consistently recognizable than others, the researchers said. For example, 99 percent of the time study participants used the same muscle movements to express happy, whereas for compound emotions, such as happily surprised, the participants expressed it the same way 93 percent of the time.

Although the collection of emotional expressions would be of most interest for basic research in cognition, researchers said that it may be also useful for understanding whether perception of compound emotions is altered in psychiatric disorders such as schizophrenia, orsocial impairments such as autism.據LiveScience網站報道，當人們感受到上述複合情感時，例如悲憤或者又開心又討厭，這些情感就會以一種特別的方式呈現在臉上。

研究人員表示，事實上，可識別的人類面部表情的數量至少是之前所認爲的三倍之多。

在一項新研究中，研究人員定義了21種用於傳遞感情的面部表情，並發現，藉助某種計算模型可以區分面部肌肉的微變化，從而準確分辨這些表情。

這些表情中，最早爲人所知的六種基本情感分別是，喜悅、悲傷、害怕、憤怒、和厭惡。另外15種則是這些基本情感的組合。例如，某人可能表現出驚喜，或者驚怒。研究人員將其統稱爲複合情感。

3月31日發表在美國國家科學院院刊上的報告顯示，對情感投射的面部表情進行這樣的詞彙擴充，有助於人腦和社會交流的研究工作、以及與人類交流的計算機系統的設計工作。

“人類的面部表情早就不侷限於表達像喜悅或者悲傷這樣的簡單感情了。我們發現，人們在運用面部肌肉表達21種表情時具有高度的一致性。”研究者亞歷克斯·馬丁內斯介紹說。他是俄亥俄州立大學電子與計算工程專業的副教授，也是一位認知學家。他說：“研究結果告訴我們，每個人在表達這21種情感時幾乎使用同一種方式，至少在我們的文化中是這樣的。”

歷史上，科學家和哲學家們都將研究重點放在六種基本情感上。但僅僅利用這六種類型就想弄清楚人腦功能是不可能的，就好比只用基本色繪製肖像畫。馬丁內斯評論說。

在他們的研究中，研究者選擇了230位大學生，要求他們對所提供的語言提示給出面部反應，然後將他們的反應通過拍照記錄下來，總共拍了大約5000張照片。提供的語言提示類似於“你剛剛得知了出人意料的好消息”（驚喜），或者“你聞到了怪味”（厭惡）。

爲了確定哪些表情足夠特別，能區分於其它表情，研究人員藉助名爲面部表情編碼系統（FACS）的計算機程序進行了數據分析。這種程序在七十年代被心理學家們廣泛應用於研究中。它將面部表情分解成小組分，例如挑眉、或者皺鼻，然後分析其中暗含的肌肉運動。正是這些肌肉運動才導致了這種表情。

FACS分析顯示，這21種表情涉及到的肌肉運動組合彼此不同，可以作爲區分依據。一種面部感知計算模型可以區分這些表情，在確定六種基本情感時準確率高達96.9%，15種複合表情的準確率達到了76.9%。

研究人員說，一些表情更容易區分。例如，99%的研究參與者使用同樣的肌肉運動表達喜悅，但對於複合情感，如驚喜，他們的一致性只有93%。

收集情感表達雖然是認知學基礎研究中很有趣的部分，但研究人員表示，它對理解複合情感的感知在精神疾病（如精神分裂），或者社交障礙（如自閉症）中是否發生變化也同樣有用。