How your face betrays your personality, health and intelligence

By David Robson

You might expect a great philosopher to look past our surface into the depths of the soul – but Ancient Greek thinkers were surprisingly concerned with appearance. Aristotle and his followers even compiled a volume of the ways that your looks could reflect your spirit.

“Soft hair indicates cowardice and coarse hair courage,” they wrote. Impudence, the treatise says, was evident in “bright, wise-open eyes with heavy blood-shot lids”; a broad nose, meanwhile, was a sign of laziness, like in cattle.

Sensuous, fleshy lips fared little better. The philosophers saw it as a sign of folly, “like an ass”, while those with especially thin mouths were thought to be proud, like lions.

Today, we are taught not to judge a book by its cover. But while it is wise not to set too much by appearances, psychologists are finding that your face offers a window on our deepest secrets. Even if you keep a stony poker face, your features can reveal details about your personality, your health, and your intelligence.

“The idea is that our biology, like genes and hormone levels, influences our growth, and the same mechanisms will also shape our character,” explains Carmen Lefevre at Northumbria University.

Consider the face’s bone structure – whether it is relatively short and wide or long and thin. Lefevre has found that people with higher levels of testosterone tend to be wider-faced with bigger cheekbones, and they are also more likely to have more assertive, and sometimes aggressive, personalities.

The link between face shape and dominance is surprisingly widespread, from capuchin monkeys – the wider the face, the more likely they are to hold a higher rank in the group’s hierarchy – to professional football players. Examining the 2010 World Cup, Keith Welker at the University of Boulder, Colorado, recently showed that the ratio of the width and height of the footballers’ faces predicted both the number of fouls among midfielders, and the number of goals scored by the forwards.

(To calculate this yourself, compare the distance from ear-to-ear with the distance between the top of your eyes, and your upper lip. The average ratio of width-to-height is around 2 – Abraham Lincoln was 1.93)

It may even clue you in to a politician’s motives. Using volunteers to rate former US presidents on different psychological attributes, Lefevre found that face shape seemed to reflect their perceived ambition and drive. John F Kennedy had a thicker-set face than 19th Century Chester Arthur, for instance. Such analyses of historical figures are perhaps best taken with a pinch of salt, however, and it has to be said that other traits, such as cooperation and intelligence, should be equally important for success.

As you might expect, your health and medical history are also written in your countenance – and the detail it offers is surprising. The amount of fat on your face, for instance, provides a stronger indication of your fitness than more standard measures, such as your body mass index. Those with thinner faces are less likely to suffer infections, and when they do, the illness is less severe; they also tend to have lower rates of depression and anxiety, probably because mental health is often closely related to the body’s fitness in general.

How could the plumpness of your cheeks say so much about you? Benedict Jones at the University of Glasgow thinks a new understanding of fat’s role in the body may help explain it. “How healthy you are isn’t so much about how much fat you have, but where you have that fat,” he says. Pear-shaped people, with more weight around the hips and bottom but slimmer torsos, tend to be healthier than “apples” with a spare tyre around the midriff, since the adipose tissue around the chest is thought to release inflammatory molecules that can damage the core organs. Perhaps the fullness of your face reflects the fatty deposits in the more harmful areas, Jones says. Or it could be that facial fat is itself dangerous for some reason.

Besides these more overt cues, very subtle differences in skin colour can also reveal your health secrets. Jones and Lefevre emphasise this has nothing to do with the tones associated with ethnicity, but barely-noticeable tints that may reflect differences in lifestyle. You appear to be in more robust health, for instance, if your skin has a slightly yellowish, golden tone. The pigments in question are called carotenoids, which, as the name suggest, can be found in orange and red fruit and veg. Carotenoids help build a healthy immune system, says Lefevre. “But when we’ve eaten enough, they layer in the skin and dye it yellow. We exhibit them, because we haven’t used them to battle illness.” The glow of health, in turn, contributes significantly to your physical attraction – more so, in fact, than the more overt tones that might accompany a trip to the tanning salon.

A blush of pink, meanwhile, should signal the good circulation that comes with an active lifestyle – and it might also be a sign of a woman’s fertility. Jones has found that women tend to adopt a slightly redder flush at the peak of the menstrual cycle, perhaps because estradiol, a sex hormone, leads the blood vessels in the cheek to dilate slightly. It may be one of many tiny shifts in appearance and behaviour that together make a woman slightly more attractive when she is most likely to conceive.

As Jones points out, these secrets were hiding in plain sight – yet we were slow to uncover them. At the very least, this knowledge helps restore the reputation of “physiognomy”, which has suffered a somewhat chequered reputation since Aristotle’s musings. Tudor king Henry VIII was so sceptical of the idea that he even banned quack “professors” from profiting from their readings, and its status took a second bashing in the early 20th Century, when it was associated with phrenology – the mistaken idea that lumps and bumps on your head can predict your behaviour.

But now the discipline is gaining credibility, we may find that there are many more surprises hiding in your selfies. Intriguingly, we seem to be able to predict intelligence from someone’s face with modest accuracy – though it’s not yet clear what specific cues make someone look smart. (Needless to say, it is not as simple as whether or not they wear glasses.) Others are examining the “gaydar”. We often can guess someone’s sexual orientation within a split-second, even when there are no stereotypical clues, but it’s still a mystery as to what we’re actually reading. Further research might explain exactly how we make these snap judgements.

It will also be interesting to see how the link between personality, lifestyle and appearance changes across the lifetime. One study managed to examine records of personality and appearance, following subjects from the 1930s to the 1990s. The scientists found that although baby-faced men tended to be less dominant in their youth, they grew to be more assertive as the years wore on – perhaps because they learnt to compensate for the expectations brought about by their puppyish appearance.

More intriguingly, the authors also found evidence of a “Dorian Gray effect” – where the ageing face began to reflect certain aspects of the personality that hadn’t been obvious when the people were younger. Women who had more attractive, sociable, personalities from adolescence to their 30s slowly started to climb in physical attractiveness, so that in their 50s they were considered better-looking than those who had been less personable but naturally prettier. One possibility is that they simply knew how to make the best of their appearance, and that their inner confidence was reflected on subtle differences in expression.

After all, there is so much more to our appearance than the bone structure and skin tone, as one particularly clever study recently demonstrated. The scientists asked volunteers to wear their favourite clothes, and then took a photo of their face. Even though the clothes themselves were not visible in the mugshots, impartial judges considered them to be considerably more attractive than other pictures of the participants. The finding is particularly striking, considering that they were asked to keep neutral expressions: somehow, the boosted self-esteem shone through anyway.

Our faces aren’t just the product of our biology. We can’t change our genes or our hormones – but by cultivating our personality and sense of self-worth, they may begin to mirror something far more important.

You are surprisingly likely to have a living doppelganger

By Zaria Gorvett

It’s on your passport. It’s how criminals are identified in a line-up. It’s how you’re recognised by old friends on the street, even after years apart. Your face: it’s so tangled up with your identity, soon it may be all you need to unlock your smartphone, access your office or buy a house.

Underpinning it all is the assurance that your looks are unique. And then, one day your illusions are smashed.

“I was the last one on the plane and there was someone in my seat, so I asked the guy to move. He turned around and he had my face,” says Neil Douglas, who was on his way to a wedding in Ireland when it happened.

“The whole plane looked at us and laughed. And that’s when I took the selfie.” The uncanny events continued when Douglas arrived at his hotel, only to find the same double at the check-in desk. Later their paths crossed again at a bar and they accepted that the universe wanted them to have a drink. He woke up the next morning with a hangover and an Argentinian radio show on the phone – the picture had gone viral.

Folk wisdom has it that everyone has a doppelganger; somewhere out there there’s a perfect duplicate of you, with your mother’s eyes, your father’s nose and that annoying mole you’ve always meant to have removed. The notion has gripped the popular imagination for millennia – it was the subject of one of the oldest known works of literature – inspiring the work of poets and scaring queens to death.

But is there any truth in it? We live on a planet of over seven billion people, so surely someone else is bound to have been born with your face? It’s a silly question with serious implications – and the answer is more complicated than you might think.

In fact until recently no one had ever even tried to find out. Then last year Teghan Lucas set out to test the risk of mistaking an innocent double for a killer.

Armed with a public collection of photographs of U.S. military personnel and the help of colleagues from the University of Adelaide, Teghan painstakingly analysed the faces of nearly four thousand individuals, measuring the distances between key features such as the eyes and ears. Next she calculated the probability that two peoples’ faces would match.

What she found was good news for the criminal justice system, but likely to disappoint anyone pining for their long-lost double: the chances of sharing just eight dimensions with someone else are less than one in a trillion. Even with 7.4 billion people on the planet, that’s only a one in 135 chance that there’s a single pair of doppelgangers. “Before you could always be questioned in a court of law, saying ‘well what if someone else just looks like him?’ Now we can say it’s extremely unlikely,” says Teghan.

The results can be explained by the famed infinite monkey problem: sit a monkey in front of a typewriter for long enough and eventually it will surely write the Complete Works of William Shakespeare by randomly hitting, biting and jumping up and down on the keys on the board.

It’s a mathematical certainty, but reversing the problem reveals just how staggeringly long the monkey would have to toil. Ignoring grammar, the monkey has a one in 26 chance of correctly typing the first letter of Macbeth. So far, so good. But already by the second letter the chance has shrunk to one in 676 (26 x 26) and by the end of the fourth line (22 letters) it’s one in 13 quintillion. When you multiply probabilities together, the chances of something actually happening disappear very, very quickly.

Besides, the wide array in human guises is undoubtedly down to more than eight traits. Far from everyone having a long-lost “twin”, in Teghan’s view it’s more likely no one does.

But that’s not quite the end of the story. The study relied on exact measurements; if your doppelganger’s ears are 59 mm but yours are 60, your likeness wouldn’t count. In any case, you probably won’t remember the last time you clocked an uncanny resemblance based on the length of someone’s ears.

There may be another way – and it all comes down to what you mean by a doppelganger. “It depends whether we mean ‘lookalike to a human’ or ‘lookalike to facial recognition software’,” says David Aldous, a statistician at U.C. Berkeley.

Francois Brunelle, who has photographed over 200 pairs of doubles for his project I’m not a look-alike, agrees. “For me it’s when you see someone and you think it’s the other person. It’s the way of being, the sum of the parts.” When seen apart, his subjects looked like perfect clones. “When you get them together and you see them side by side, sometimes you feel that they are not the same at all.”
If fine details aren’t important, suddenly the possibility of having a lookalike looks a lot more realistic. But is this really true? To find out, first we need to get to grips with what’s going on when we recognise a familiar face.

Take the illusion of Bill Clinton and Al Gore which circulated the internet before their re-election in 1997. It features a seemingly unremarkable picture of the two men standing side by side. On closer inspection, you can see that Gore’s “internal” facial features – his eyes, nose and mouth – have been replaced by Clinton’s. Even without these traits, with his underlying facial structure intact Al Gore looks completely normal.

It’s a striking demonstration of the way faces are stored in the brain: more like a map than an image. When you bump into a friend on the street, the brain immediately sets to work recognising their features – such as hairline and skin tone – individually, like recognising Italy by its shape alone. But what if they’ve just had a haircut? Or they’re wearing makeup?

To ensure they can be recognised in any context, the brain employs an area known as the fusiform gyrus to tie all the pieces together. If you compare it to finding a country on a map, this is like checking it has a border with France and a coast. This holistic ‘sum of the parts’ perception is thought to make recognising friends a lot more accurate than it would be if their features were assessed in isolation. Crucially, it also fudges the importance of some of the subtler details.

“Most people concentrate on superficial characteristics such as hair-line, hair style, eyebrows,” says Nick Fieller, a statistician involved in The Computer-Aided Facial Recognition Project. Other research has shown we look to the eyes, mouth and nose, in that order.

Then it’s just a matter of working out the probability that someone else will have all the same versions as you. “There are only so many genes in the world which specify the shape of the face and millions of people, so it’s bound to happen,” says Winrich Freiwald, who studies face perception at Rockefeller University. “For somebody with an ‘average’ face it’s comparatively easy to find good matches,” says Fieller.

Let’s assume our man has short blonde hair, brown eyes, a fleshy nose (like Prince Philip, the Duke of Edinburgh), a round face and a full beard. Research into the prevalence of these features is hard to come by, but he’s off to a promising start: 55% of the global population has brown eyes.

Meanwhile more than one in ten people have round faces, according to research funded by a cosmetics company. Then there’s his nose. A study of photographs taken in Europe and Israel identified the ‘fleshy’ type as the most prevalent (24.2%). In the author’s view these are also the least attractive.

Finally – how much hair is there out there? If you thought this was too frivolous for serious investigation, you’d be wrong: among 24,300 people surveyed at a Florida theme park, 82% of men had hair shorter than shoulder-length. Natural blondes, however, constitute just 2%. As the ‘beard capital’ of the world, in the UK most men have some form of facial hair and nearly one in six have a full beard.
A simple calculation (male x brown eyes x blonde x round face x fleshy nose x short hair x full beard) reveals the probability of a person possessing all these features is just over one in 100,000 (0.00001020%).

That would give our guy no less than 74,000 potential doppelgangers. Of course many of these prevalence rates aren’t global, so this is very imprecise. But judging by the number of celebrity look-alikes out there, it might not be far off. “After the picture went viral I think there was a small army of us at some point,” says Douglas.

So what’s the probability that everyone has a duplicate roaming the earth? The simplest way to guess would be to estimate the number of possible faces and compare it to the number of people alive today.
You might expect that even if there are 7.4 billion different faces out there, with 7.4 billion people on the planet there’s clearly one for everyone. But there’s a catch. You’d actually need close to 150 billion people for that to be statistically likely. The discrepancy is down to a statistical quirk known as the coupon collector’s problem. Let’s say there are 50 coupons in a jar and each time you draw one it’s put back in. How many would you need to draw before it’s likely you’ve chosen each coupon at least once?

It takes very little time to collect the first few coupons. The trouble is finding the last few: on average drawing the last one takes about 50 draws on its own, so to collect all 50 you need about 225. It’s possible that most people have a doppelganger – but everyone? “There’s a big difference between being lucky sometimes and being lucky always,” says Aldous.

No one has any good idea what the first number is. Indeed, it may never be possible to say definitively, since the perception of facial resemblance is subjective. Some people have trouble recognising themselves in photos, while others rarely forget a face. And how we perceive similarity is heavily influenced by familiarity. “Some doubles when they get together, they say ‘No I don’t see it. Really, I don’t.’ It’s so obvious to everyone else; it’s a little crazy to hear that,” says Brunelle.
Even so, Fieller thinks there’s a good chance. “I think most people have somebody who is a facial lookalike unless they have a truly exceptional and unusual face,” he says. Friewald agrees. “I think in the digital age which we are entering, at some point we will know because there will be pictures of almost everyone online,” he says.

Why are we so interested anyway? “If you meet someone that looks like you, you have an instant bond because you share something.” Brunelle has received interest from thousands of people searching for their lookalikes, especially from China – a fact he puts down to the one-child policy. Research has shown we judge similar looking-people to be more trustworthy and attractive – a factor thought to contribute to our voting choices.

It may stem back to our deep evolutionary past, when facial resemblance was a useful indicator of kinship. In today’s globalised world, this is misguided. “It is entirely possible for two people with similar facial features to have DNA that is no more similar than that of two random people,” says Lavinia Paternoster, a geneticist at the University of Bristol.

And before you go fantasising about doing a temporary life-swap with your ‘twin’, there’s no guarantee you’ll have anything in common physically either. “Well I’m 5’7 and he’s 6’3… so it’s mainly in the face,” says Douglas.

Facial structure predicts goals, fouls among World Cup soccer players

World Cup soccer players with higher facial-width-to-height ratios are more likely to commit fouls, score goals and make assists, according to a study by a researcher at the University of Colorado Boulder.

The structure of a soccer player’s face can predict his performance on the field—including his likelihood of scoring goals, making assists and committing fouls—according to a study led by a researcher at the University of Colorado Boulder.

The scientists studied the facial-width-to-height ratio (FHWR) of about 1,000 players from 32 countries who competed in the 2010 World Cup. The results, published in the journal Adaptive Human Behavior and Physiology, showed that midfielders, who play both offense and defense, and forwards, who lead the offense, with higher FWHRs were more likely to commit fouls. Forwards with higher FWHRs also were more likely to score goals or make assists.

“Previous research into facial structure of athletes has been primarily in the United States and Canada,” said Keith Welker, a postdoctoral researcher in CU-Boulder Department of Psychology and Neuroscience and the lead author of the paper. “No one had really looked at how facial-width-to-height ratio is associated with athletic performance by comparing people from across the world.”

FWHR is the distance between the cheekbones divided by the distance between the mid-brow and the upper lip. Past studies have shown that a high FWHR is associated with more aggressive behavior, with both positive and negative results. For example, high FWHR correlates with greater antisocial and unethical behavior, but it also correlates with greater success among CEOs and achievement drive among U.S. presidents. However, some previous research has failed to find a correlation between FWHR and aggressive behavior in certain populations.

The new study adds weight to the argument that FWHR does correlate with aggression. Welker and his colleagues chose to look at the 2010 World Cup because of the quality and quantity of the data available. “There are a lot of athletic data out there,” Welker said. “We were exploring contexts to look at aggressive behavior and found that the World Cup, which quantifies goals, fouls and assists, provides a multinational way of addressing whether facial structure produces this aggressive behavior and performance.”

Scientists have several ideas about how FWHR might be associated with aggression. One possibility is that it’s related to testosterone exposure earlier in life. Testosterone during puberty can affect a variety of physical traits, including bone density, muscle growth and cranial shape, Welker said.

Co-authors of the study were Stefan Goetz, Shyneth Galicia and Jordan Liphardt of Wayne State University in Michigan and Justin Carré of Nipissing University in Ontario, Canada. –

See more at: