Face facts: our first impressions are wrong

Alexander Todorov, a professor of psychology at Princeton University and one of the world’s experts on the subject, has spent years researching our bent towards facial stereotyping and has now written a mesmerising book on the subject called Face Value: The Irresistible Influence of First Impressions.

He conducted a famous study in which subjects were shown a series of photographs, each containing a pair of faces, and were asked to choose based purely on instinct which one looked more competent. The images were of leading candidates in state governor or Senate elections elsewhere in the US, so the subjects did not recognise them. When they compared their choices with the poll results they found the subjects had predicted the winners in 72.4 per cent of the senatorial races and 68.6 per cent of the gubernatorial ones. In a nutshell, “politicians who look more competent are more likely to win elections”.

So, I ask him, what does “competence” tend to look like. The answer isn’t great news for the shy, plain female. The main three components tend to be a) attractiveness, b) a masculine face and c) a confident look. So, as Todorov says, “it would favour attractive men who appear confident”. Feminine faces are seen as more trustworthy. Yet when it comes to politicians it seems we value competence over trustworthiness.

We also trust “typical” faces — ie ones that aren’t unusual. At first sight, he says, “we like and trust people who resemble people who we already like and trust”. If your father is a warm person you will instinctively assess people who resemble him as being warm too, even though you could be completely wrong and probably are. And we prefer faces that reflect our own. “Nice” people look like us. “There is a little Narcissus in each of us,” writes Todorov. Tests show we are more willing to invest money in people and to vote for politicians whose faces have been manipulated to resemble our own. There is also “evidence that we are more likely to marry people who look like us and even choose purebred dogs who ��resemble’ us for pets”. Indeed, Face Value includes amusing pictures of human beings and their “lookalike” dogs.

The book is full of photographic experiments that prove that the tiniest alterations to a person’s face — to their nostril, upper lip, eyes — can change what we think about them. One asks you to consider two faces and decide which would make the better leader. Liberal voters, it transpires, tend to pick one face and conservatives the other (I, it seems, am a liberal). The preferences, says the book, reflect our ideological stereotypes of right-wing, masculine, dominant-looking leaders and left-wing, feminine, non-dominant-looking leaders. It also proves the importance of eyebrows. In a fascinating set of photographs, one shows Richard Nixon without eyes and one with eyes but without eyebrows. Without eyes you can still tell it is President Nixon. Without his eyebrows but with his eyes he is unrecognisable. Yet how many love songs have been written about eyebrows?

The trouble with our snap facial judgments is that they are very unreliable. Todorov cites a study in which experienced personnel managers and social workers, whom you might assume to be good judges of character, were asked to judge 150 photos of students on intelligence. The correlations of the judgments with the measured intelligence of the students and their grades was barely above zero. In other words, they couldn’t predict their intelligence.

However, they agreed with one another in their assessments. Those judged to be intelligent had “typical” facial features, pleasant expressions and more neat appearances. Another study found that faces deemed to show traits such as friendliness, kindness and honesty tended to have similar features — shallow-set eyes, light eyebrows, light complexion and medium-width face.

“Baby-faced” adults are, apparently, perceived as being naive, submissive, kind and warm. In The New York Times last month Jason Dana, an assistant professor of management and marketing at the Yale School of Management, wrote that job interviews are utterly useless at identifying the best candidate because interviewers “typically form strong but unwarranted impressions about interviewees, often revealing more about themselves than the candidates”.

“So great is people’s confidence in their ability to glean valuable information from a face to face conversation,” he wrote, “that they feel they can do so even if they know they are not being dealt with squarely. But they are wrong.”

As Todorov says: “When it comes to character, faces provide very weak signals.” In the book he adds: “Evolution has equipped us with a readiness to attend to faces. This readiness develops into an intricate network of brain regions specialised to process faces. These networks support our extraordinary face skills. But these skills ironically support the illusion of accuracy of first impressions. Both the skills and the illusion make us believe that faces provide a wealth of information about the person even when they don’t.”

So why do we put so much store by first impressions when they are not accurate? Todorov says that in most cases we probably don’t find out. You’ll make a snap judgment of someone behind a shop counter who may simply be having a bad day or have a headache and never see them again, so your assessment isn’t challenged. Sometimes we do find out we’re wrong. When Warren Harding became the 29th US president, physiognomists saw signs of presidential greatness in his face, with a forehead indicating “broad-mindedness and intellectual powers” and “strong will-power and great endurance combined in one chin”. As it turns out, historians agree that Harding was the worst American president, his administration becoming best known for scandals involving bribery and incompetence.

However, “face reading” has been irresistible to humans for a long time. The first preserved document dedicated to physiognomy is Physiognomica, a treatise attributed to Aristotle, which posits that the character of animals is revealed in their form and humans resembling them possess the character of these animals. This theory was developed in the 16th century by Giovanni Battista della Porta, an Italian scholar, who produced a book contending, for instance, that people who look like cows (big foreheads, wide noses) are stupid, lazy and cowardly.

In the 19th century Cesare Lombroso, the founding father of criminal anthropology, wrote books on how villains can be identified by physical characteristics. In Criminal Man he suggested that they were evolutionary degenerates, closer to primates than humans, and that the typical criminal had “jug ears, thick hair, thin beard, pronounced sinuses, protruding chin and broad cheekbones”. Murderers had glassy eyes and big, hawk-like noses, while rapists “nearly always have sparkling eyes, delicate features and swollen lips and eyelids”.

We may laugh at such nonsense now, but the facial stereotyping we continue to practise can have serious consequences. Two British researchers studied faces from real police line-ups. When participants were asked to pick the likely culprit they tended to choose the “criminal-looking” faces. Analysis showed the line-ups were biased: the police officers had selected face foils who were less criminal-looking than the suspect.

Recently two psychologists tested whether impressions of trustworthiness predict death sentences in Florida. Defendants who looked more untrustworthy were more likely to get the death sentence. A follow-up study looked at innocent people wrongly convicted of murder and subsequently exonerated. Again, the unlucky souls who were perceived as less trustworthy were more likely to have been wrongfully sentenced to death. Todorov reminds us that in the Middle Ages there was a law by which when two people were suspected of a crime, the uglier was selected for punishment. We may not have moved on as much as we think.

Some “face reading” has had great value over the centuries. It is very useful, for instance, to detect when someone is angry or very sick. However, the idea that faces are a map of the personality is an illusion. Indeed, the impressions we draw reveal our own biases and stereotypes. A good example of taking action to minimise such bias, he says, is the increase in the number of women in philharmonic orchestras. Until relatively recently they were heavily populated by men. The introduction of “blind auditions”, where the judges could only hear the candidate’s performance, not see it, meant they couldn’t be influenced by appearances.

When it comes to politics, the candidate’s appearance seems to affect only those who know nothing about politics and some swing/undecided voters. Todorov says what appearance-influenced voters are doing is substituting a hard decision with an easy one. “Finding out whether a politician is truly competent takes effort and time. Deciding whether a politician looks competent is an extremely easy task. Appearance-influenced voters are looking for the right information in the wrong place because it is easy to do so.”

It would be impossible in practice, but theoretically would it be a more satisfactory state of affairs if we only heard the politicians explaining their policies and did not see them? “Perhaps,” he says, “but people draw the same kind of inferences from voice information too. Just from a single word — hello — people make inferences.”

Perhaps, with our own general election happening today, we should be thankful that we are not one of those countries that includes the candidates’ photos on the ballot paper.
Face Value: The Irresistible Influence of First Impressions by Alexander Todorov is published by Princeton, £27.95