冰鑑: 以貌取人

清代, 曾國藩以相貌論人優劣, 並著有《冰鑑》一書以論相學; 《清史稿·曾國藩傳》載:「國藩為人威重, 美須髯, 目三角有棱。每對客, 注視移時不語、見者悚然, 退則記其優劣, 無或爽者。」。人的相貌多少會被心思意念影響。一個人品性的大概, 或可由容貌觀之。對面目猙獰、眼神兇惡者, 自然先察後親; 眉清目秀、氣宇軒昂者, 人多先親而後察, 戒備之心自然較少。然而, 又說「仗義每多屠狗輩，負心盡是讀書人。」; 若以相貌為斷, 不加細察, 便淪為「以貌取人」、有過之而不及了。以「相學」卜卦、故弄玄虛以為趨吉避凶者, 實為「混飯吃」的街頭小學, 更不足道。

雜誌《經濟學人》報導一個研究項目, 證實50個白種男性商人(Fortune 1000 表列的首末各25位)的容貌可傳達有關其能力、可親度、誠信等的資訊。節錄如下:

Physiognomy and success

Face value

Jan 24th 2008
From The Economist print edition

What the boss looks like determines how he performs

A COUPLE of years ago a group of management scholars from Yale and the University of Pittsburgh tried to discover if there was a link between a company's success and the personality of its boss. To work out what that personality was, they asked senior managers to score their bosses for such traits as an ability to communicate an exciting vision of the future or to stand as a good model for others to follow. When the data were analysed, the researchers found no evidence of a connection between how well a firm was doing and what its boss was like. As far as they could tell, a company could not be judged by its chief executive any better than a book could be judged by its cover.

A few years before this, however, a team of psychologists from Tufts University, led by Nalini Ambady, discovered that when people watched two-second-long film-clips of professors lecturing, they were pretty good at determining how able a teacher each professor actually was. At the end of the study, the perceptions generated by those who had watched only the clips were found to match those of students taught by those self-same professors for a full semester.

Now, Dr Ambady and her colleague, Nicholas Rule, have taken things a step further. They have shown that even a still photograph can convey a lot of information about competence—and that it can do so in a way which suggests the assessments of all those senior managers were poppycock.

Dr Ambady and Mr Rule showed 100 undergraduates the faces of the chief executives of the top 25 and the bottom 25 companies in the Fortune 1,000 list. Half the students were asked how good they thought the person they were looking at would be at leading a company and half were asked to rate five personality traits on the basis of the photograph. These traits were competence, dominance, likeability, facial maturity (in other words, did the individual have an adult-looking face or a baby-face) and trustworthiness.

By a useful (though hardly unexpected) coincidence, all the businessmen were male and all were white, so there were no confounding variables of race or sex. The study even controlled for age, the emotional expression in the photos and the physical attractiveness of the individuals by obtaining separate ratings of these from other students and using statistical techniques to remove their effects.

This may sound like voodoo. Psychologists spent much of the 20th century denigrating the work of 19th-century physiognomists and phrenologists who thought the shapes of faces and skulls carry information about personality. However, recent work has shown that such traits can, indeed, be assessed from photographs of faces with a reasonable accuracy.

And Dr Ambady and Mr Rule were surprised by just how accurate the students' observations were. The results of their study, which are about to be published in Psychological Science, show that both the students' assessments of the leadership potential of the bosses and their ratings for the traits of competence, dominance and facial maturity were significantly related to a company's profits. Moreover, the researchers discovered that these two connections were independent of each other. When they controlled for the “power” traits, they still found the link between perceived leadership and profit, and when they controlled for leadership they still found the link between profit and power.

These findings suggest that instant judgments by the ignorant (nobody even recognised Warren Buffett) are more accurate than assessments made by well-informed professionals. It looks as if knowing a chief executive disrupts the ability to judge his performance.

Sadly, the characteristics of likeability and trustworthiness appear to have no link to company profits, suggesting that when it comes to business success, being warm and fuzzy does not matter much (though these traits are not harmful). But this result also suggests yet another thing that stockmarket analysts might care to take into account when preparing their reports: the physog of the chief executive.