They found that people made judgements about the attractiveness, likeability, trustworthiness, competence and aggressiveness of other people after looking at their faces for 100 milliseconds.

Rather than correcting initial judgements, longer glances at the same faces caused people to become more convinced of their initial opinions, the scientists found.

"The link between facial features and character may be tenuous at best, but that doesn't stop our minds from sizing other people up at a glance," Alex Todorov, an assistant professor of psychology, said.

"We decide very quickly whether a person possesses many of the traits we feel are important, such as likeability and competence, even though we have not exchanged a single word with them. It appears we are hard-wired to draw these inferences in a fast, unreflective way."

The researchers showed photographs of faces to 200 participants and asked them to rate the images according to how attractive, likeable, trustworthy, competent or aggressive the people seemed.

There was no significant change between the opinions people formed after one-tenth of a second and those they formed after half a second or a whole second.

"What we found was that, if given more time, people's fundamental judgment about faces did not change," Mr Todorov said. "Observers simply became more confident in their judgments as the duration lengthened."

He suggested the speed of judgement of trustworthiness could reflect activity in an area of the brain connected with feelings of fear.

Full approval for the project was given on 22 August by the UK Medical Research Council and the Wellcome Trust, which are funding the £61-million project. It follows the success of a three-month pilot project in Manchester involving 3800 participants, which received glowing reviews from an independent international panel.

Each volunteer participating will donate small samples of urine and blood, containing their DNA, for indefinite storage in the "bank", which will be based in Manchester. They will also respond to detailed questionnaires about their lifestyle, health and environment.

By following all participants until their death, researchers hope to identify the genetic and lifestyle factors which may have contributed to any illnesses they suffered. If genes linked to disease can be identified, it might be possible to prevent illness in carriers of the gene by altering their lifestyles, for example.
Vital insights

Information from the project may also lead to development of new drugs and treatments, the researchers hope.

"For decades to come, the UK Biobank resource should provide researchers around the world with vital insights into some of the most distressing diseases of middle and old age," says Rory Collins, Biobank's principal investigator.

Eligible citizens throughout the UK will receive invitations to participate from early 2007. The organisers plan to recruit the full 500,000 over the next three or four years at 35 temporary centres to be set up throughout the UK.
Sceptical outlook

But despite its grandeur, the project has its critics. "For the millions of people thinking about whether or not to give their DNA, many questions remain unanswered," says Helen Wallace at GeneWatch, an independent watchdog in Buxton, Derbyshire, monitoring potentially controversial developments in genetics.

Wallace also doubts whether the information on lifestyle will be detailed enough - such as on diet - to validate disease links drawn with people's genes. "We remain sceptical whether the Biobank can give meaningful results when many exposures will not be measured accurately and most genes will have small and variable effects," she says.

Collins says these criticisms are unfair. "It's the largeness and the richness of the data that will make Biobank unique," he told New Scientist.
Tough questions

"The questionnaire is the most difficult part," Collins explained, because unlike the stored DNA and fluid samples which can be revisited time and time again, the baseline information can only be collected once. "You have to ask all the questions that people might have wished had been asked in 10 years time and beyond," he says.

To address that, the questionnaire is very detailed, with hundreds of questions. In the pilot phase, Biobank researchers in Manchester successfully tested out a 30-minute to 40-minute touch-screen method to collect personal data.

Collins said that safeguards were in place to make sure that the sensitive personal information would be kept secret, and that the information would be made anonymous before release to researchers, making it impossible to trace the medical information back to an identifiable individual.
Gold standard

"Identities would be available only to a small number of senior staff at Biobank to enable baseline data to be linked to a patient's medical records held by the UK National Health Service," says Collins. "It will not be available outside, and information will not be fed back to volunteers, their family doctors or their families."

The review panel, led by Thorkild Sørensen at the Institute of Preventive Medicine in Copenhagen, Denmark, said the measures for ethical oversight and governance of the samples were "exemplary, and would be held up as a gold standard across the world".

A similar project is under development in the US backed by Francis Collins, director of the National Human Genome Research Institute in Bethesda, Maryland, but is some way behind the UK Biobank.

CRAIG VENTER can make a unique claim. "To my knowledge I am the only person on the planet to have their genome decoded," he says with a hint of satisfaction. He is the driving force behind the commercial effort to sequence the human genome, so it is perhaps not surprising he was the first to attain this level of self-knowledge.

He hopes he will not be alone for long. The J. Craig Venter Science Foundation is sponsoring a $10 million prize for anyone who can get the cost of sequencing an individual's genome down to $1000. Venter predicts it will be scooped within a decade. At this price, plenty of people will be able to unravel their personal genetic story. "That's when we're going to see a massive shift in the study of human genetic variation and our own personal genetic make-up," he says.

Venter believes that once geneticists can compare thousands or even millions of individual genomes, they will get a better handle on why, as individuals, we turn out the way we do. Already we can link certain genes with diseases, personality traits, cognitive styles, aptitudes and the like. And although Venter does not believe in genetic determinism - the view that our fundamental nature is written in our genes - he still champions the importance of genes. "We're clearly not born blank slates, but I think the genetic effects on these traits are a lot stronger than most people are willing to deal with." Venter thinks that the analysis of entire personal genomes, together with information about how people actually live their lives, will provide a whole new window onto the complex interplay between your nature and your nurture. This avalanche of genomic knowledge could render the Freudian revolution a footnote of intellectual history. And unlike psychoanalysis, it might even give you the information you need to ensure your kids reach their full potential.

Some readers will find this prospect tremendously exciting, others quite terrifying. The fact is, though, that if Venter's dream is ever to become reality, a lot of people need to get their DNA sequenced. To get the ball rolling, Venter plans to unveil his genomic portrait on GenBank, the US National Institutes of Health genetic sequence database, making it freely available to anyone. Perhaps you are tempted to join him. If so, before you shell out the cash, you will probably want to know what you will be getting for your money. Put plainly, will you attain a level of self-knowlege previously unknown to humankind, or will your genome tell you little you didn't already know?
Do you really want to know?

The most obvious benefit of genomic knowledge is in predicting if and when you are likely to succumb to illnesses such as heart disease, stroke, Parkinson's and diabetes. Armed with this information, you could decide to take preventive medicines, or make lifestyle changes to delay or stave off the development of a condition. In a decade, we will undoubtedly know far more about how certain versions of various genes are linked to particular conditions. Venter, unsurprisingly, is ahead of the game. A look through his genome has prompted him to start taking cholesterol-lowering drugs called statins to delay or prevent heart disease.

Such insight sounds attractive, but it also raises the thorny issue of how to deal with information about your chances of future health, particularly as medical professionals might not be on hand to provide their interpretive and counselling skills. In a few cases the implications of finding a certain mutation will be certain. People with a particular duplicated region of chromosome 4, for example, always develop Huntington's disease. More often, though, the link between a sequence of DNA and any disease it might contribute to is much less clear-cut, often depending on the combined effects of many genes and environmental factors.

This goes to the heart of the main problem with personal genomics. Gone are the days when people talked about "the gene for X". Rare cases like Huntington's aside, at best all we can say about most sequences of DNA is that they confer on carriers a certain likelihood of developing a particular condition or trait. For example, a genetic variant could have a harmful effect in 70 per cent of people, but no detectable effect in the remaining 30 per cent. "Understanding this whole field will come down to probability statistics, which are very hard to interpret," admits Venter. In essence, the vagaries of genetic prediction mean that some people might benefit greatly, and others not at all.

The problem of probabilities is even greater when trying to understand the genetic basis of aspects of personality, temperament and intelligence than for many health-related traits. The details of human behaviour are not spelled out in the genome; social and cultural factors profoundly influence how we think and act. That doesn't render genes irrelevant. Genes are crucial to the development of the mind, personality and behaviour, just as they are for building limbs, lungs and livers. "Genes help wire up brains and guide individual neurons to their destination," says Gary Marcus, a developmental psychologist at New York University. "So genes play a pretty big role in building some of our cognitive processes, like the kinds of memory and fundamental reasoning skills we have."

In the absence of any other clues about a person's aptitudes, genetic analysis, though imperfect, might be better than nothing. "Good teachers try to tailor their teaching to an individual's strengths and weaknesses," says Marcus. "The genome could be one more tool to guess what these are." Genetic forecasting about talents or deficits that are hard to assess early on in life could also prove useful. "Such predictions might be useful if they are made in childhood. Parents can make extra efforts, for example, to encourage the musical development of a child with 'musicality' genes," says Judith Rich Harris, an independent psychologist based in New Jersey.

This is already possible to some extent. "There are other ways of finding out that one has untapped talents," Harris points out. As a child she disliked and avoided maths, but in college she did surprisingly well on a maths aptitude test, and even enjoyed taking some maths classes. Later, when bedridden through illness, she worked out a mathematical theory of human information processing that has since been published. "Doing well on that test at the age of 20 made me realise that my earlier avoidance of math had been a mistake," Harris says.

Venter's biography offers a similar example. After enlisting in the US navy, he, like all new recruits, went through a series of examinations, including an IQ test. When the results came back, Venter found himself top out of 35,000 test scores. This served as a wake-up call, making him reconsider his academic potential, and setting him on the path to his present position.

Clearly, life offers us many chances to test our aptitudes, and information from a personal genome could be a useful complement. It might even provide a basis for making ball-park predictions about how children will turn out. As adults, though, we don't just want to know about our strengths and weaknesses, we also want to understand how we have become what we are. Where some people now look to psychotherapists for such answers, in the future others may be tempted to consult their DNA. Could the information packed into our genomes really change our view of ourselves as individuals?

The answer might seem to be a resounding "yes". After all, geneticists already compare the composite human and chimp genomes to get insights into the genetic basis of chimp and human natures. So shouldn't a comparison of genomic portraits from two individuals reveal the genetic basis of why they differ?
What genes won't tell you

This is tricky terrain. The first misstep is to unreasonably elevate the power of genes. "We measure the genetic effects because we can," Venter says. "We don't know how to measure all the environmental effects that impact on our lives, or we would. The danger lies in attributing too much significance to something just because you can measure it."

A second and bigger stumbling block is that when thinking about individuals, probability complicates matters enormously. "Probabilities are perfectly reasonable as summaries of tendencies of multiple events or individuals in a large sample," says Steven Pinker, a cognitive psychologist at Harvard University, "but they are puzzling, perhaps even meaningless, when applied to a single event or individual." The relevance of probabilities after the event is even more questionable. It is a moot point, for example, whether gene analysis could be used objectively to determine the likelihood that a given pianist would have become a pianist.

All this is an attempt to explain retrospectively why you are the way you are. Though you can assign probabilities to candidate causes - to this gene or that environmental effect - either, or both, could have played a part in shaping you, with plenty of other factors and some random noise thrown in for good measure.
Your long-lost twin

Nevertheless, there is a way of thinking about your genomic portrait that might go some way towards helping you come to terms with who you are. "Having one's genome decoded is like finding out that you have an identical twin who was reared somewhere else," says Harris. "It lets you see the possibilities your genes provided you with, the selves you might have become, but didn't."

The "genomic twin" idea gives a visceral edge to thinking about the cold, abstract strings of DNA code that spell out our genomes. Harris tells a true story of twins separated at birth and raised apart. One became a concert pianist, the other was offered music lessons in childhood but declined. "I wonder what that twin thought when she met her concert-pianist sister," she says. "Did she realise that she, too, might have been an excellent musician if she had only accepted the offer to take lessons?" These are ultimately unanswerable speculations, but that doesn't diminish their intrigue for such curious animals as humans.

Now imagine that you have had your genome decoded, and the sequence has been sent to you on a CD. You pop it in your computer, run a scan and find out that you have "musicality" genes (imagining for the moment they exist and have been identified). How would you react? Well, such a discovery should prompt the same kinds of reflections suggested by Harris, with the same caveat. In both cases, we cannot unravel the specific genetic, environmental and random effects that made us the way we are - the probabilities make the picture so fuzzy you can't make out details. In this case, the best you can do is to try your hand at piano and see how it goes. Maybe you'll pick it up, maybe you won't.

In many cases, knowing that you have gene variants associated with certain strengths or weaknesses, tastes or temperaments will probably add little to your self-knowledge. "If I know that I'm impulsive, that should affect how I evaluate an appealing temptation in front of me now. If I know I have genes that probabilistically predispose me to impulsiveness, that is far less relevant to my situation," Pinker argues. "It's the impulsiveness that matters - whether it was ultimately due to genes, environment, chance events in zigzagging neurons is far less important, indeed, perhaps irrelevant."

What this boils down to is that if you are looking for self-knowledge, genes are often not the best level of analysis. And that's not the worst of it. "One of the biggest dangers of the information in the genome is over-interpretation by society," says Venter. Genes at best provide possibilities, not certainties. A particular suite of genes might tend to be associated with high intelligence or a love of risks, but they are no guarantee of such traits. The fact that we can talk about such genes does not mean we should attribute too much power to them. Another is the spectre of self-fulfilling prophecies. If you know that you have genes typically associated with being bad at music, you might convince yourself that you are just born without musicality. "If people expect that they're not good at something, they tend to become not good at that thing," says Marcus.

Looking into your genome could also dredge up trouble. Imagine that a genome scan revealed that a law-abiding man had genes that gave him a 50 per cent chance of becoming a criminal. "Will he feel good that he isn't in fact a criminal, or will the information itself make it more likely that he will embark upon a new career as a law-breaker?" asks Harris. "And what about the person with a similar genome who has already committed a serious crime? Can we blame him and punish him for his crime, if it was his genes that made him do it?"

With all this in mind, what do the experts conclude about the wisdom of getting our genomes sequenced as soon as the technology makes it feasible? Venter is optimistic about the promise of greater knowledge. At the moment we face difficult questions about the underlying causes for our unique repertoire of character traits. "When we have millions of genomes in hand we'll begin to be able to answer these questions," says Venter. "Today we know almost nothing compared with what we'll know in a decade or so."
Get one for your kids

Others are more circumspect, though there is general agreement that a genomic portrait would provide useful medical forecasts and other practical insights. Whether it would help you redefine your self-image is more contentious. "For an adult, I don't think it would make much difference. Whatever the genome brings us has already been considerably shaped by the environment, enough so that you can learn a lot more about a person by studying her directly than by studying her genome," says Marcus. "But new parents might some day want to purchase them to get an early forecast of their children's likely aptitudes. After all, even before we start to talk we've got a genome."

As the physicist Niels Bohr famously quipped: "Prediction is very difficult, especially about the future." One thing is certain, though: the more we study human genetic variation, the more we will know about the genetic basis of our predispositions, predilections and potential, so boosting our ability to predict how genes will affect the way people turn out. Some of this progress will come through comparative studies of the composite human genome and the genomes of chimps and other primates. More will result from targeted studies into particular diseases and conditions. And, if enough people follow Venter into the realm of personal genomics and make their individual DNA available for study, this will undoubtedly provide an additional and powerful lens through which to look at nature and nurture, and how they modulate each other.

The genome will never tell you everything, however. Even with a better understanding of which genes are important and why, our genes will never define our destiny. Venter is living proof of this. By many measures he is a risk-taker - he is an avid sailor and has taken big scientific chances in his career - but an analysis of his genome for genes associated with risk-taking and thrill-seeking failed to turn up the expected variants. That doesn't make him any less of a risk-taker. It just means that his personality has developed that way regardless of what his genes say.

From issue 2565 of New Scientist magazine, 19 August 2006, page 28-36