SOLAR power has always had a reputation for being expensive, but not for much longer. In India, electricity from solar is now cheaper than that from diesel generators.

The news - which will boost India's "Solar Mission" to install 20,000 megawatts of solar power by 2022 - could have implications for other developing nations too.

Recent figures from market analysts Bloomberg New Energy Finance (BNEF) show that the price of solar panels fell by almost 50 per cent in 2011.

They are now just one-quarter of what they were in 2008.

That makes them a cost-effective option for many people in developing countries.

A quarter of people in India do not have access to electricity, according to the International Energy Agency's 2011 World Energy Outlook report.

Those who are connected to the national grid experience frequent blackouts.

To cope, many homes and factories install diesel generators.

But this comes at a cost.

 Not only does burning diesel produce carbon dioxide, contributing to climate change, the fumes produced have been linked to health problems from respiratory and heart disease to cancer.

Now the generators could be on their way out. In India, electricity from solar supplied to the grid has fallen to just 8.78 rupees per kilowatt-hour compared with 17 rupees for diesel.

The drop has little to do with improvements in the notoriously poor efficiency of solar panels: industrial panels still only convert 15 to 18 per cent of the energy they receive into electricity.

But they are now much cheaper to produce, so inefficiency is no longer a major sticking point.

It is all largely down to economies of scale, says Jenny Chase, head of solar analysis at BNEF. In 2011, enough solar panels were produced worldwide to generate 27 gigawatts, compared with 7.7 GW in 2009.

 Chase says solar power is now cheaper than diesel "anywhere as sunny as Spain".

That means vast areas of Latin America, Africa and Asia could start adopting solar power.

"We have been selling to Asia and the Middle East," says Björn Emde, European spokesman for Suntech, the world's largest producer of silicon panels.

 Over the next few years he expects to add South Africa and Nigeria to that list.

The one thing stopping households buying a solar panel is the initial cost, says Amit Kumar, director of energy-environment technology development at The Energy and Resources Institute in New Delhi, India.

Buying a solar panel is more expensive than buying a diesel generator, but according to Chase's calculations solar becomes cheaper than diesel after seven years.

The panels last 25 years.

Even in India, solar electricity remains twice as expensive as electricity from coal, but that may soon change.

While the price drop in 2011 was exceptional, analysts agree that solar will keep getting cheaper.

Suntech's in-house analysts predict that, by 2015, solar electricity will be as cheap as grid electricity in half of all countries.

When that happens, expect to see solar panels wherever you go.

A Russian drilling team is trying to confirm that they have finally hit Lake Vostok, a vast subglacial body of water hidden 3.5 kilometres beneath the surface of the Antarctic ice sheet

A spokesperson for the Russian Antarctic Expedition in St Petersburg told New Scientist this morning that the drill made contact with water late last week and then automatically withdrew up the borehole, as planned.

That suggests the lake has been breached, but the team are now checking the level of water in the borehole and readings from pressure sensors to confirm that the water did come from the lake and not a pocket of water in the ice above the lake.

 Ice temperatures rise as you go deeper into the ice sheet, and approach melting point just above the lake, so the fact that the team hit liquid water doesn't necessarily mean they've reached the lake.

"For the time being we are waiting for official confirmation," said the spokesperson. An announcement is expected within the next two days.

No more drilling

Drilling stopped on 5 February and most of the team, led by Valerii Lukin, have left the area.

Two team members have remained to monitor the borehole over the Antarctic winter.

Even if Lukin's team have broken through the ice sheet to the lake, they will still need to wait nearly a year to sample its secrets.

To avoid contaminating Vostok with drilling fluid Lukin and his team planned from the start to pierce the roof of the sealed ice cave which encases the lake and then let pressure in the lake force water into the drill hole.

The plan is to leave the lake water to freeze in the borehole and create a plug, preventing contamination.

The team will return to sample it during the following austral summer.

Life, or nothing

Lake Vostok has been isolated from the surface for millions of years, and many hope it contains bizarre new life forms.

At present, however, that seems unlikely.

The drillers have already sampled wedges of accretion ice – lake water that has naturally frozen onto the underside of the ice sheet – and although some researchers claim it contains bacteria, others write this off as contamination.

Moreover, the ice above is loaded with bubbles of trapped air.

That air has accumulated in the lake for millennia, boosting the oxygen concentrations in the water and creating a potentially toxic environment. Some say that as a result, it is likely that the lake is completely sterile.

That could be just as interesting.

If Lake Vostok turns out to be sterile, that will make it the only place on Earth where there is water but no life.

Gabrielle Walker is the author of Antarctica: An intimate portrait of the world's most mysterious continent, to be published by Bloomsbury on 1 March

Editorial: "No room for complacency over solar storms"

THE sun is gearing up for a peak in activity at a time when technology makes our planet more vulnerable to solar outbursts than ever before.

 Monitoring has improved since the last solar maximum, so what are the big risks this time around?

About once every 11 years, the sun goes ballistic, throwing out more bursts of magnetic activity than normal.

As a large but harmless solar flare signalled last week, the next solar maximum is due in 2013.

In the past, these storms have triggered extra currents in power lines, destroying transformers and leading to blackouts.

 This time around, blackouts could be more common.

John Kappenman of Storm Analysis Consultants in Duluth, Minnesota, found that many transformers in the US are ageing and therefore extra fragile.

He also points out that while new transformers consume less power, that means relatively small currents from solar storms can overload and damage them.

 "If anything, we're making things on the grid more vulnerable," he says.

As well as causing blackouts, solar storms can fry satellite electronics, which we rely on more and more for communication, navigation and weather forecasts.

To assess how vulnerable this leaves us, New Scientist enlisted the help of the Union of Concerned Scientists and Jonathan McDowell of the Harvard-Smithsonian Center for Astrophysics, both based in Cambridge, Massachusetts, who each keep careful records of satellite launches and failures.

They calculated that there are 994 working satellites in orbit today compared with 629 during the sun's last peak. Better storm forecasting should make them less vulnerable (see "And now for the solar forecast"). Ground controllers can command satellites to switch off sensitive parts temporarily in response to a forecast.

However, there is another risk that barely existed 11 years ago.

Many passenger flights between North America and Asia now take shortcuts over the North Pole.

This saves flying time and cuts fuel consumption, but it leaves planes vulnerable to solar storms.

Earth's magnetic defences are weakest at the poles (see diagram), allowing electrons and protons to pour into the atmosphere during solar storms.

This can interfere with planes' communication and navigation signals.

Airlines reroute polar flights when solar storms are predicted, as they did last week, adding hours of flight time and costing tens of thousands of dollars in extra fuel per flight.

Astronauts on the International Space Station, meanwhile, get an extra dose of radiation during a storm: there are six there now compared with three 11 years ago.

We can take some comfort in the knowledge that the looming maximum is supposed to be relatively weak, but we shouldn't be complacent. In 1859, during an otherwise weak cycle, a solar storm made telegraph wires spark, starting fires.

"You've got the opportunity for flares, and they can be big ones," warns David Hathaway of NASA's Goddard Space Flight Center in Greenbelt, Maryland.

Newt Gingrich described himself as a visionary when he unveiled plans to create a mammoth new space programme, including a permanent colony on the moon within the next nine years.

Within eight years, he pledges a new Mars rocket programme – specifically, a "continually operating propulsion system capable of getting to Mars within a remarkably short time".

He also reiterated his plan to declare at least part of the moon as US territory, with colonists capable of petitioning for statehood status.

There is little doubt that Gingrich believes in big ideas. Unfortunately, however, there is a difference between big ideas and good ideas.

After all, being a visionary doesn't mean abandoning practicality altogether but rather harnessing it creatively to make new things happen.

Put aside that Gingrich was speaking in Florida, the state most invested in space exploration and, by happenstance, the next up on the Republican primary schedule. Let's consider cost first.

The Apollo missions to the moon cost in excess of $100 billion in current dollars. In 2005, NASA administrator Michael Griffin estimated the cost of a programme to land four astronauts on the moon by 2018 (as was then planned), at $104 billion.

Who will pay?

Now, four astronauts is not a permanent colony on the moon.

To have a permanent colony, you would have to manufacture housing, most likely underground, or at least under significant shielding, since there is no atmosphere and no magnetic field to shield against the harmful effects of cosmic rays for an extended period.

Not to mention the need to build facilities for waste recycling, plus food storage and preparation.

That is, unless we continually provide food and other provisions for pilgrims from Earth, creating a non-self-sustaining colony.

But Gingrich has already made it quite clear, in his attacks on President Obama, that he would not like to be remembered for championing any such sort of government-sponsored food programme.

So, to truly embark on such an endeavour within a decade, we would have to spend somewhere between a few hundred billion and a trillion dollars.

Whether we could develop the necessary technology for such a task within a decade is an open question, although for a sufficiently large investment, it might not be impossible.

However, Gingrich is vying for leadership of a party whose major rallying cry is an end to big government programmes and make-work projects to stimulate the economy.

Gingrich might argue that we need not rely on government for the investment.

However, without a clear business plan, it is hard to imagine private money investing $1 trillion in a programme with no clear commercial goal.

Yet he did not explain precisely what he wanted to do with such a colony, or what it might achieve, besides potentially populating a new 51st state.

Certainly the goal would not be a scientific one, since there is little scientific gain to be made that would justify the cost, and one could populate the whole solar system with unmanned spacecraft that could explore all the planets and their moons for this cost, as well as send up satellites that could map the heavens on unprecedented scales.

Business inopportunity

So is manufacturing his goal?

But what would we manufacture on the moon that we could not do on Earth for a fraction of the cost?

It is true that there may be significant amounts of terrestrially rare isotopes like helium-3 in the lunar soil, and some have argued that this would be useful for fusion power here on Earth.

But since we don't yet know how to produce fusion power on Earth, it seems a little premature to rush out on a trillion-dollar adventure to gather up potential fuel.

Perhaps we could put mirrors on the moon to beam sunlight to Earth for power.

But given that currently 10,000 times the total energy used by humanity on a daily basis falls on the Earth from the sun, it is not clear that we need to go to the moon to harness more of it.

Gingrich also said during this same address that he envisions a vibrant commercial near-Earth space programme for the purposes of science, tourism and manufacturing. Once again, he didn't bother to explore precisely what sort of programme one might envisage here.

It took more than $100 billion to manufacture a white elephant in near-Earth orbit called the International Space Station, a large, smelly metal can that to date has produced no science, no manufacturing and tourism that only billionaires could afford.

Perhaps Gingrich imagines a vibrant Earth-surveying programme that might help monitor climate change?

No, probably not.

Reality suspended

Not content to merely colonise the moon in a decade, Gingrich has also promised to develop a viable Mars programme to begin human space exploration of that planet within the next decade.

It is hard to imagine why he didn't also promise an intergalactic starship in this timeframe as well, as long as he was being visionary.

Finally, Gingrich may not be aware that the current US flags on the moon don't mean the US owns it, any more than those on US research stations in Antarctica mean the US owns that continent.

But I suppose if one is willing to suspend reality to imagine creating an imaginary new expensive, and expansive, space programme from nothing in a mere decade, without raising the taxes to do it, anything is possible.

 It certainly seems easier to imagine populating the moon in this way than actually solving the very real problems we face on Earth today.

This article originally appeared in Slate. Lawrence Krauss is foundation professor and director of the Origins Project at Arizona State University in Tempe. His newest book is A Universe from Nothing: Why there is something rather than nothing

YOU'RE running late for work and you can't find your keys.

What's really annoying is that in your frantic search, you pick up and move them without realising.

This may be because the brain systems involved in the task are working at different speeds, with the system responsible for perception unable to keep pace.

So says Grayden Solman and his colleagues at the University of Waterloo in Ontario, Canada.

To investigate how we search, Solman's team created a simple computer-based task that involved searching through a pile of coloured shapes on a computer screen.

Volunteers were instructed to find a specific shape in a stack as quickly as possible, while the computer monitored their actions.

"Between 10 and 20 per cent of the time, they would miss the object," says Solman, even though they picked it up.

"We thought that was remarkably often."

To find out why, the team developed a number of further experiments.

To check whether volunteers were just forgetting their target, they gave a new group a list of items to memorise before the search task, which they had to recall afterwards.

The idea was to fill each volunteer's "memory load", so that they were unable to hold any other information in their short-term memory.

Although this was expected to have a negative effect on their performance at the search task, the extra load made no difference to the percentage of mistakes volunteers made.

To check that the volunteers were paying enough attention to the items they were moving, Solman's team created another task involving a stack of cards marked with shapes that only became visible while the card was being moved.

Again, they were surprised to see the same level of error, says Solman.

Finally, the team analysed participants' mouse movements as they were carrying out a similar search task.

 They discovered that volunteers' movements were slower after they had moved and missed their target (Cognition, DOI: 10.1016/j.cognition.2011.12.006).

Solman's team propose that the system in the brain that deals with movement is running too quickly for the visual system to keep up.

While you are rummaging around a messy house to find your keys, you might not be giving your visual system enough time to work out what each object is.

Since time can be costly, sacrificing accuracy on occasion for speed might be beneficial overall, Solman thinks.

The slowing of mouse movements suggests that at some level the volunteers were aware that they had missed their target, a theory that is backed up by other studies that show people tend to slow down their actions after they have made a mistake, even if they don't consciously realise the mistake.

Solman reckons this reflects the brain's "attempt to slow down the motor system", to allow the visual system to catch up and conscious perception to occur.

"What's really interesting is the notion that the motor and perceptual system are decoupled.

They're both trying to help you find [your keys] but they're not coordinating," says Todd Horowitz, at Harvard University.

"There are implications for social search, such as a doctor looking through an X-ray or [security] looking through luggage."

Venezuela's president has mused that the US may have given him and other leftist leaders cancer. T

hat's unlikely to be possible

Five South American presidents and former presidents, including Venezuela's Hugo Chavez, have been recently diagnosed with cancer.

Chavez speculated that US agents may be inducing the disease in South American leaders by feeding them or injecting them with an unspecified substance. The state department has rejected Chavez's insinuation.

Can you give someone cancer?

Not reliably.

Injecting cancerous cells into a person isn't enough to give him the disease.

The abnormal tissue has to penetrate and grow in other areas of the body.

If you injected someone with live cancer cells, their immune system would almost certainly attack and destroy the foreign tissue. In theory, secret agents might be able to induce cancer in a leftist South American president with a severely weakened immune system.

Or perhaps they could harvest tissue from him, expose it to a carcinogen, and then reintroduce it into his body.

As far as we know, however, these techniques have never successfully caused cancer in a human.

While it's tough to induce cancer in an enemy, it's certainly possible to increase his chances of developing the disease. The most effective option would be radiation.

Oncologists implant radiation-emitting devices the size of a seed into some patients to combat existing cancers. It's hard to say just how much the device would increase a healthy individual's risk of cancer, but leaving a high-intensity model inside the body for weeks or months would result in a significant dose of radiation.

The victim would likely notice the implant, though.

They're too big for an ordinary needle, and need to be inserted through a catheter.

You could, alternatively, contaminate the victim's diet with high levels of aflatoxin, which is associated with liver cancer.

Or you could infect her or him with any of a number of cancer-causing biological agents. Helicobacter pylori contributes to the development of gastric cancer, and human papillomaviruses can cause cervical, anal and a few other forms of cancer.

But these tactics probably wouldn't produce cancer in the short term and aren't guaranteed to have any effect at all. In countries with high aflatoxin exposure, like China and parts of Africa, fewer than 1 in 1000 people develop liver cancer.

Most of the research on infusing cancer into humans is decades old. In the 1950s, Chester Southam gained notoriety by injecting hundreds of cancer patients and healthy prison inmates with live cancer cells.

Southam wasn't trying to give his subjects cancer.

Rather, he was testing the efficiency with which the patients' immune systems would reject the cells.

He was so confident that the patients would fight off the invaders that he thought it unnecessary to tell them what he was doing.

None of Southam's patients seems to have developed metastatic cancer from his injections, and most modern oncologists believe the experiment posed little risk to the subjects.

(One of the patients showed signs of a potentially spreading disease before dying of a separate illness.)

Southam was sanctioned for fraudulent practices, however, and the case helped establish modern informed consent standards.

Southam's experiments were abandoned in the 1950s, but he wasn't the last doctor to inject a patient with live cancer cells.

In 2009, a Taiwanese doctor was accused of implanting cancerous uterine cells into healthy patients as part of an insurance scam.

While the insurance companies were out more than $660,000, none of the victims developed cancer.

Today, ethical physicians inject live cancer cells only into laboratory animals such as mice and rats.

In most cases, the animals' immune systems are compromised, or the rodents have been genetically engineered to rapidly spread mutant cells

It's a way off, I know, but don't bother to order a diary for 2013 – you won't be needing one. On 21 December 2012, the world as we know it will come to an end.

It's all been revealed in an old calendar of the Mayans, if you believe how it's interpreted by various self-appointed experts on their slightly whacky websites.

The coming Armageddon could, it's said, take many forms: a disastrous surge in solar activity, a reversal of Earth's magnetic poles, a collision with a black hole, the close passage of a mysterious planet called Nibiru – the list goes on. It's all very entertaining, and pure nonsense.

The Mayans did keep a calendar, called the "long count", based on a period of 1,872,000 days.

It began in August 3114 BC, and so is due to click over to its next cycle late next year.

However, there's no evidence to suggest the Mayans saw the switch from one cycle to the next as apocalyptic. More to the point, even if they did, they had no way of foretelling the future.

On the other hand, it's true that we are doomed.

Our planet and everything on it won't last. A billion years from now the sun will have brightened and swollen to a point at which the oceans will start to evaporate.

A billion years after that, all Earth's surface water will be gone and, with it, all life except for some hardy species that can survive on whatever moisture remains underground.

Incoming!

The finite lifespan of the sun guarantees the demise of planet three.

But there are all kinds of other natural calamities that might kill off large swathes of us in the much shorter term – and we love to talk about them.

Asteroid or comet collisions are a big favourite.

The Earth has repeatedly been used for target practice by big dumb objects in the past.

An asteroid at least 10 kilometres across barrelled into us just over 65 million years ago and helped wipe out the last of the dinosaurs along with many other groups of animals and plants.

A good thing, too, if you're human, because it left mammals free to flourish.

In 1908 a blast with the strength of about a thousand Hiroshima atomic bombs flattened trees over a wide area around the Tunguska river in Siberia.

The presumed culprit in this case was a fragment of a comet or a large meteoroid that exploded several kilometres above the surface. Had it happened over a populous city, the effect would have been disastrous.

Other nasty stuff has happened to the Earth.

Supervolcanoes have erupted, blanketing vast areas with dust and lava, and plunging the globe into deep volcanic winters. Ice ages have come and gone, and very occasionally, during "snowball Earth" events, it seems that almost the entire planet has frozen over for millions of years.

These kind of events will happen again.

We will be hit by asteroids and comets, large and small.

More supervolcanoes will erupt.

It's inevitable, and some of these events are perfectly capable of decimating the human race, or even driving us to extinction virtually overnight.

Chances are…

The trouble is, the level of danger and the immediacy of the threat is often grossly overstated.

Run-ins with asteroids as big as that which put paid to the dinosaurs happen once every 100 million years or so. But that doesn't mean they happen every 100 million years like clockwork.

Much smaller impacts, like the Tunguska episode, occur on average about once every thousand years – a problem if they happen to maliciously target built-up areas, but not so alarming when you think of the great tracts of ocean and wilderness which are much more likely to be on the receiving end.

Speculation about upcoming disasters runs rampant when there's a failure, or unwillingness, to grasp basic facts.

It happened this year in connection with an innocuous little comet called Elenin.

Google its name and you'll find all kinds of hair-raising stories about how Elenin would come close to or even ram into the Earth, bringing death and destruction on a biblical scale.

The hysteria started shortly after the comet's discovery when a few armchair theorists mistook the size of Elenin's coma – the glowing, almost vacuum-thin shiny fuzz of vaporised particles around the hard nucleus – for the size of the nucleus itself.

Word quickly spread on the internet, helped by the usual eagerness of tabloids and late-night chat shows to exploit a juicy yarn, that Elenin was as big as a planet and would cause chaos during its close passage of the Earth.

In fact, as astronomers knew, Elenin was modest by cometary standards and never going to come any closer than 35 million kilometres, or about 90 times as far away as the moon. In the event, it disintegrated and was lost from view to even the most powerful telescopes.

Muddled science opens the door to tales of Armageddon. But it's helped along by the fact that we love a scary story, for the same reason we enjoy a good horror or sci-fi flick – because it lets us escape the banality of everyday life.

 For some, there's another reason to suspend disbelief – the expectation that some time soon the day of judgement will be upon us.

The 2012 phenomenon is alive and kicking, and will doubtless remain so until 22 December next year, when we wake up to find Earth hasn't been knocked off its axis or suffered any other ill effects from obscure cosmic alignments, geomagnetic reversals or close fly-bys of unknown planets.

If past experience is anything to go by, explanations of how we survived will quickly emerge, and new doomsday forecasts will replace the ones that didn't quite work out.

As always, these revised warnings will find an eager audience.

And, as always, the scientific reality will take a back seat.

All scientific research funded by British taxpayers will be made available online free of charge, according to a government report published earlier this week.

And it doesn't stop there – the government intends the website, to be named Gateway to Research, to eventually incorporate research funded by other bodies.

Much of the high-energy physics research community currently uses a system of open-access online publishing, and Janet Finch, former vice-chair of Keele University, UK, has been charged with investigating how the UK might set up something similar for all its taxpayer-funded science.

The arxiv.org website, an online repository set up in 1991, offers almost all high-energy physics research for free. Despite this, established physics journals have reported no decrease in subscriptions.

The announcement is part of a growing trend towards open access, following the success of the free journals PLoS One and, more recently, Nature Scientific Reports.

No way to get a job

Critics of the government's announcement have argued that these prominent open-access journals already allow researchers to publish their research online for free, and yet some still choose not to. Björn Brembs of the Free University Berlin, Germany, a prominent advocate of open access to research, says this criticism is unfair: "One of the major obstacles to open-access publishing is that the highest-ranking journals have chosen not to go open-access. We may in principle be free to publish wherever we want already, but not if we want to get a job."

British science minister David Willetts suggests that peer-reviewed journals could become open-access by charging their contributors, rather than their readers. "One of the clear options is to shift from a system in which university libraries pay for journals to one in which the academics pay to publish," he says. "But then you need to shift the funding so that the academics could afford to pay to publish."

However, Brembs argues that paying to publish won't tackle one of the major issues that sparked the open-access movement – the high price of journals. "It's the spiralling charges by the publishers which really need be brought under control."

Neuroscience is becoming increasingly relevant to the law. Can brain scans ever prove whether you are guilty or innocent?

NEUROSCIENTISTS seek to understand how the brain underpins our behaviour, thoughts and feelings.

 Given that the law is also concerned with human behaviour, albeit for quite different reasons, it is hardly surprising that remarkable advances in our understanding of the brain have led many to believe that neuroscience is becoming increasingly relevant to the law.

In the US, a number of universities teach courses on the interface between neuroscience and the law, and the Chicago-based MacArthur Foundation has invested several million dollars to fund research in this area.

In the UK, the Royal Society has just published a report on neuroscience and the law.

Some argue that neuroscience has already cast doubt on the idea of free will, and therefore raises questions about the legitimacy of punishing people for actions over which they had no control.

 In the US there is a steady increase in defence attorneys seeking to introduce neuroscientific evidence. So is "my brain made me do it" a legitimate defence in a criminal trial?

There will surely be cases where such evidence is relevant. Most countries specify an age of criminal responsibility somewhere between 6 and 16; in England and Wales it is 10. Brain imaging studies have shown that the brain continues to develop throughout adolescence, with the prefrontal cortex, implicated in impulse control and decision-making, not reaching maturity until 20 or so. Such studies have also shown that there are huge individual differences. It is hard to believe that all 10-year-olds should be held fully responsible when they break the law.

Few situations are so clear-cut, however. For example, brain imaging has shown that there are often differences between the brains of people categorised as psychopaths or with antisocial personality disorder (who are disproportionately likely to commit violent crimes) and the more law-abiding majority. But it needs to be stressed that these are average differences only - it is not possible to diagnose individual criminal psychopathy on the strength of a brain scan.

In any case, if psychopaths think, feel and behave differently from others, then of course these differences will be reflected in their brains. That is hardly sufficient to establish that the atypical nature of their brain was the cause of their behaviour.

Similarly, men who were abused as children and who also carry a particular version of a gene called monoamine oxidase-A (MAOA) are more likely to behave violently. But not all men with the gene and a history of childhood abuse commit violent crimes; in one study fewer than a third had been convicted of one by the age of 26 (Science, vol 297, p 851). Should a convicted criminal's sentence be reduced on the strength of a genetic screen?

Rather than such evidence serving to reduce a convicted criminal's sentence, one could argue that it might be used to increase it, or at least influence decisions about release or parole. Such decisions involve assessing the risk of reoffending, and such risk assessment is notoriously imprecise.

Unsurprisingly, those who have to make these decisions err on the side of caution. In 2003 the UK introduced indeterminate sentencing, which allowed judges not only to set a minimum prison sentence but also to require convicts to satisfy the authorities that they would not pose any threat if released. The provisions were initially designed to detain a small number of exceptionally dangerous criminals, but by March 2011 there were 6550 people in prison under these terms. Even if indeterminate sentences were abolished, the problem of risk assessment would not go away. It seems at least possible that neuroscientific or genetic evidence might be able to reduce the risk of getting these decisions wrong.

The law is also concerned with establishing whether people are telling the truth. Is the witness who claims to have seen the defendant at the scene of the crime lying? Is the defendant who protests his innocence actually guilty? It is widely recognised that polygraphs are not reliable enough to be used in a court of law. Might fMRI brain scans, which can detect changes in brain activity when people perform a particular task, do a better job? Several experiments have shown differences in activity in certain regions of the brain when people are asked to answer some questions truthfully and others with a lie.

Two companies, Cephos and No Lie MRI, have already been set up in the US to commercialise these discoveries. Neither has yet succeeded in persuading a court to accept their evidence - and there are good reasons to remain sceptical. At best fMRI might sometimes be able to detect a difference between a witness who is telling the truth and another who deliberately lies. But there is good experimental evidence to suggest that brain imaging will not distinguish between a reliable witness and one who is honestly mistaken. By the same token it seems probable that defendants who have repeatedly protested their innocence under persistent questioning might end up believing they are telling the truth even if they are not. There is also evidence that people can be taught to fool the machine.

There's no doubt neuroscience will provide some startling revelations about human behaviour in the coming years, but we must not get ahead of ourselves. At this point, our priority needs to be to make sure that advances that may affect the law are communicated properly to legal professionals so that when it becomes appropriate, neuroscience is used in court to the benefit of all involved.

Nicholas Mackintosh is professor emeritus in the department of experimental psychology at the University of Cambridge and chair of the working group for the Royal Society's report Brain Waves Module 4: Neuroscience and the law

Science thrives on freedom of expression and must be at the forefront of defending it

THE words "science" and "censorship" do not sit easily together.

And yet over the past decade, science has come to occupy an increasingly important role in debates over free speech.

This is partly due to public clashes between science and politics, from the censoring of climate science in the US under the Bush administration to David Nutt's dismissal as the UK government's adviser on drugs after voicing his views on the safety of ecstasy.

But it also reflects a revolution in access to information which has exposed every sector of society to an unprecedented level of scrutiny.

From WikiLeaks to phone hacking, the tension between openness, privacy and confidentiality has become one of the defining issues of our time.

Scientists have unexpectedly found themselves at the heart of this debate, as the latest round of leaked climate emails makes abundantly clear.

In recognition of this trend, the award-winning magazine Index on Censorship, which explores challenges to freedom of speech, has dedicated its latest issue, "Dark Matter", to science.

One well-documented clash between science and censorship is in the use of libel actions to try to silence scientists and science writers; the journal Nature and Richard Dawkins are among the most recent to face suits.

Scientists and science writers have emerged from some of these battles as free speech champions and martyrs, notably the writer Simon Singh, cardiologist Peter Wilmshurst and NASA climate scientist James Hansen.

There have also been striking incidents within science itself, perhaps most notoriously during the original "climategate" scandal at the Climate Research Unit of the University of East Anglia in Norwich, UK.

 The hacked emails revealed a reluctance to comply with freedom of information requests and possible attempts to conceal data.

The information commissioner recently ruled that UEA should release its data, and partly in response to climategate, the UK's Royal Society has set up an investigation into openness.

Not surprisingly there are debates about the proper course of action.

Our special issue explores two opposing views. Fred Pearce, the leading chronicler of climategate, makes the argument for open access for the benefit of science and public discourse.

Michael Halpern of the US Union of Concerned Scientists warns about freedom of information being deployed as a form of harassment. He is calling on legislators to consider whether there is sufficient protection of academic free speech.

This view has been echoed in the UK by Royal Society president Paul Nurse, as well as in the House of Lords during a debate on the proposed Protection of Freedoms legislation.

The bill includes an amendment to the Freedom of Information Act which will oblige public authorities to release data sets in reusable electronic form and extend the range of FOI to the wider public sector.

Two of the academics in the Lords, historian Paul Bew and philosopher Onora O'Neill, raised concerns about the consequences for research.

Bew has suggested including an exemption for unpublished research (which already exists in Scottish FOI legislation), warning of the possible harm that may be caused if data is released before it has been peer-reviewed.

However, even if an exemption is included in the bill, the combination of hackers, leakers and the sheer momentum of the open-access movement is likely to limit its scope, particularly for politically sensitive research.

The leak of a further 5000 climategate emails last week is a case in point. So there may be no other choice but to embrace full transparency.

Any discussion about access to information cannot ignore the suppression of data within the drugs and medical devices industry.

Lack of transparency in drug trials has left doctors dangerously ignorant of potential side effects.

This is nothing new, but the demand for openness here too may become irresistible.

As Deborah Cohen reports in our issue, Thomas Jefferson of the Cochrane Collaboration believes that open access should be the default setting for drug trials once a drug is registered.

Yet despite the backing of all the most eminent scientific institutions for openness there has been limited success.

For now the focus remains on libel.

The pressing need for reform has resulted in an unprecedented campaigning alliance between free speech groups and science.

For the past two years, the organisation Index on Censorship has been working on this with Sense about Science and the writers' association English PEN.

There is no doubt that libel's chilling effect on scientific research and discourse has been a pivotal factor in the success of the campaign.

While politicians are suspicious of giving any further freedom to the media, when presented with evidence of the extent to which scientists and science writers have been silenced and bullied by individuals, interest groups and industry, they have found it impossible to ignore.

Reform that makes it less easy to use the law as a tool of intimidation and that introduces a robust public interest defence will be of critical importance for the future open discussion of issues of scientific concern.

As Wilmshurst and Singh have demonstrated in their own costly and exhausting libel battles, all too often the fight for free speech depends on the courage of individuals.

Both the law and the culture within the science establishment have to change in order to safeguard open debate. Freedom of expression depends on it.

Jo Glanville is editor of Index on Censorship, the magazine of the London-based campaigning organisation of the same name.

Index on Censorship is hosting a debate called Is transparency bad for science? at Imperial College London on 6 December.

To attend email eve@indexoncensorship.org

The phrase "shark attack" is sensationalist and damaging – bites by sharks are often investigatory or defensive

I HAVE seen a shark attack, and it is dramatic.

 On a recent trip to Cape Town, South Africa, I watched great whites breaching from beneath seals in successful and unsuccessful predations.

These are attempts to kill.

Shark predations on seals are attacks, because the intent is clear.

However, to suggest that shark-on-human encounters should be called predations would be wrong.

The way sharks encounter seals is fundamentally different from how they treat humans.

I believe the time is right for science to reconsider its use of the phrase "shark attack" on humans.

Such language creates a one-dimensional perception of these events and makes protecting threatened shark species more difficult.

 After all, why care about an animal that wants to eat us?

Historically, the language has been less emotive.

Cases of "shark bite" were noted by doctors in 1899 and "shark accident" was an accepted term until the 1930s, even if it was fatal.

This faded when Australian surgeon Victor Coppleson concluded in a 1933 article in the Medical Journal of Australia that "the evidence sharks will attack man is complete".

The first New Orleans Shark Symposium in 1958 cemented "attack" language in the scientific community.

The argument for change is compelling. Modern research has shown that bites by sharks are often investigatory or defensive, taking place in cloudy water and out of curiosity.

Human-shark encounters are always called attacks even when there is no contact, artificially amplifying the numbers.

What's more, no distinction is made for minor bites from non-threatening species.

 In Australia, 13 per cent of all "attacks" come from small wobbegong sharks, which bite when stepped on.

Under the existing system, the public is unable to tell scratches from fatalities, boats from people or wobbegongs from great whites.

Finally, "attack" terminology creates an inappropriate connection between scientific reasoning and tabloid journalism.

 If there is no separation between science and sensationalism, then educating the public about true shark behaviour is more difficult.

Changing terminology is not simple.

 I use the phrase "shark bite incidents" but the transition cannot be unilateral.

This norm-setting task is the responsibility of the whole scientific community.

Christopher Neff is studying the politics of shark bites at the University of Sydney, Australia

Our pursuit of naught provides profound insights into the nature of reality

Read more: "The nature of nothingness"

SHAKESPEARE had it right, even in ways he couldn't have imagined.

 For centuries, scientists have indeed been making much ado about nothing - and with good reason.

Nothing, or rather what we've long taken to be nothing, may be the key to understanding everything from why particles have mass to the expansion of the universe.

 As explored in this special issue of New Scientist (see "The nature of nothingness"), nothing is a rich and subtle subject whose biography is far from finished.

The modern story of nothing began with a thought experiment dreamed up by Isaac Newton.

Imagine two identical rocks, tied together with a string, whirling around their common centre.

The string pulls taut.

But, Newton asked, how would we explain the taut string if the rocks were spinning in an otherwise empty universe?

Since motion is relative, and the rocks aren't moving relative to one another, what sets the benchmark for their motion?

 Newton concluded that the answer had to be, well, nothing: the string pulls taut because the rocks are moving relative to empty space itself.

With that answer, Newton made something out of nothing.

Since then, science has taken Newton's lead and run with it, big time.

 A century ago Albert Einstein was puzzled by gravity.

How does the sun, separated from us by 150 million kilometres, keep the Earth tethered in orbit?

His answer? Another resounding nothing. Einstein realised that in response to the presence of matter and energy, space (and time) develops warps that guide the motion of objects, like the Earth, passing through it.

Gravity, Einstein concluded, is the shape of what most people would casually think of as nothing.

The discovery of quantum mechanics took the story of nothing further still.

A hallmark of the theory is the uncertainty principle, which established that there are features of the microworld - such as the position and speed of a particle - that cannot be simultaneously determined.

Instead, such physical characteristics are subject to "quantum fluctuations" - unavoidable jitters during which their values undulate wildly.

When applied to a region of space that by any intuitive, classical measure would be deemed empty, such quantum fluctuations ensure that particles pop in and out of existence and fields fluctuate frantically.

 And this activity can be measured.

Place two metal plates close together in otherwise empty space and an imbalance in microscopic jitters outside and between the plates forces them together: nothing can make objects move.

This year's Nobel prize in physics recognises the power of nothing on cosmic scales.

In 1998, two teams of astronomers announced, shockingly, that the expansion rate of space is increasing over time.

Because of the gravitational pull of each galaxy on all others, the expansion was expected to slow.

But the prizewinning data point to an unseen energy permeating space, called "dark energy", that yields a repulsive gravitational push.

 Its identity remains a mystery, and a big one at that.

Dark energy - what would be left after removing all galaxies, stars and particles, nothing, in common parlance - accounts for more than 70 per cent of the mass of the universe.

The Large Hadron Collider near Geneva, the world's most powerful particle accelerator, is also in the business of probing nothing.

A prime goal is to find evidence for the Higgs field, which, much like dark energy, is believed to permeate empty space.

Instead of driving the expansion of space, the Higgs would exert a drag on particles, giving them the masses they have.

According to theory, slamming protons together at enormous speed should produce a piece of the Higgs field - a Higgs particle, in fact. In effect, through the violent collisions, researchers are trying to chip off a tiny piece of nothing.

Since the time of Newton, we have thus gradually realised that nature has masked the identity of nothing with a Shakespearian deftness.

With the relentless rise of science, we have slowly peeled back the obscuring layers, revealing vital intangibles at the very heart of reality, a grand triumph for nothing.

Trees may get their beautiful shapes from battling the elements.

A mathematical model shows that the pattern some branches make, first noted by Leonardo da Vinci, is the best at withstanding gusts of wind.

Da Vinci observed that at any height above the ground, the total cross section of some trees' branches has roughly the same area as that of the trunk.

This pattern was thought to accommodate the tree's plumbing, as water flows fastest when the branched pipes can hold as much water as the original pipe.

But Christophe Eloy at the University of California in San Diego thought trees contained too little plumbing to be the reason behind the pattern.

Instead he thought wind might play a role.

So he built a model to simulate the bending forces exerted by the wind, and found that trees with branch thicknesses fitting da Vinci's rule resisted breakage.

The work will appear in Physical Review Letters.

The model could help architects design wind-resistant buildings that mimic tree branches, says plant biophysicist Karl Niklas at Cornell University in Ithaca, New York.

Reference: arxiv.org/abs/1105.2591

Meat is bad: bad for you, bad for the environment. At least, that's the usual argument.

Each year, the doors to the UN climate negotiations, which kick off again in Durban, South Africa, on 28 November, are assailed by demonstrators brandishing pro-vegetarian placards.

The fact is that livestock farming accounts for a whopping 15 per cent of all greenhouse gas emissions.

We can't all go veggie, so just how much meat is it OK for an eco-citizen to eat?

It's not just the demonstrators who are concerned about food's impact on the climate.

This week, a major report concludes that food production is too close to the limits of a "safe operating space" defined by how much we need, how much we can produce, and its impact on the climate.

Meat is a major contributor to that: 80 per cent of agricultural emissions come from meat production, and the problem is getting worse.

As people get richer, the demand for protein gets stronger, says Molly Jahn, a former undersecretary at the US Department of Agriculture, and one of the authors of Achieving Food Security in the Face of Climate Change, commissioned by the Consultative Group on International Agricultural Research (CGIAR).

It's unrealistic to expect everyone to give up meat entirely, and many of the world's poor need to increase their meat consumption to overcome malnutrition and food insecurity.

The solution is to eat less meat rather than no meat.

In 2007, Colin Butler of the Australian National University in Canberra estimated that the average person consumed 100 grams of meat a day, or about one burger (a quarter-pounder is 113 g).

The rich eat 10 times more than the poor – in other words, some people get 10 burgers a day while others get none.

Butler showed that if every person in the world ate 50 g of red meat and 40 g of white meat per day by 2050, greenhouse gas emissions from meat production would stabilise at 2005 levels – a target cited in national plans for agricultural emissions.

That's about one burger and one small chicken breast per person every two days (The Lancet, DOI: 10.1016/S0140- 6736(07)61256-2).

Butler's 2007 figures didn't take into account the fact that we throw out a lot of the animal mass produced because we consider it inedible.

Western countries are the biggest offenders: while many cultures are not fazed by a meal of brains or testicles, Butler estimates that Americans and Australians throw out up to half the cow mass they produce.

At New Scientist's request, he updated his calculations.

He estimates that globally we discard between 5 and 10 per cent of the animal.

This means we can only allow ourselves 80 to 85 g of red and white meat, or one burger and one chicken fillet every three days.

That's an upper limit. Emissions may need to be cut further.

Our allowance would drop further if more people were as wasteful as the Americans and Australians.

And, according to CGIAR, in addition to the waste between the abattoir and the plate, one-third of all produced food is spoiled because of poor refrigeration, pests and bulk packaging that encourages consumers to buy more than they can eat. All of which eat into our meat allowance

Editorial: "Aspirin, the cancer wonder drug"

People at high risk of cancer may soon be advised to take readily available drugs such as aspirin to reduce their chances of succumbing to one of the world's biggest killers.

Although cancer screening programmes already exist, offering women regular smear tests or mammograms, for example, to detect early signs of cervical or breast cancer, these look for precancerous changes to cells or suspicious lumps, rather than identifying people who are at high risk of cancer in the future.

For many of these people, even those who possess a gene mutation that puts them at high risk, watchful waiting is the norm.

That could be about to change.

It looks as if common drugs may be able to slash a person's chances of developing cancer – dubbed chemoprevention.

 "For people at high risk of cancer at least, chemoprevention is finally coming of age," says John Burn of Newcastle University, UK.

Breast cancer set the trend. Women over the age of 50 are often offered mammograms to detect early signs of cancer.

 Such screening has drawn controversy, as it can flag up harmless lumps as cancerous, leading women to undergo unnecessary investigation.

However, mounting evidence suggests mammograms of healthy breasts might provide vital information on a woman's cancer risk in future, and that this information is not being put to good use.

"All the routine mammogram does is look for early cancers," says Jack Cuzick of the Wolfson Institute of Preventative Medicine in London.

"But within this mammogram there's a lot of information about who is at risk."

 What's more, tamoxifen, a cheap drug that is already used to treat breast cancer, could significantly reduce the risk of the disease developing in the first place.

Several groups have found that healthy women with dense tissue in 75 per cent or more of the breast – around 5 to 10 per cent of the female population – were around four times as likely to develop breast cancer within 10 years following the diagnosis.

Breast density relates to the amount of connective and glandular tissue in the breast, and this produces hormones that can encourage cells to divide.

"We think that this combination creates an environment in which changes are more likely to occur that can give rise to cancer in the future," says Norman Boyd of the Ontario Cancer Institute in Toronto, Canada.

Now, Cuzick and his colleagues have shown that treating women at high risk with tamoxifen can reduce breast density, cutting their risk of developing the most common form of breast cancer by up to 63 per cent.

The results were presented at the Frontiers of Cancer Prevention Research meeting in Boston last week.

Tamoxifen does have some side effects, but for women whose mammograms suggest that they are at high risk, it could be an attractive option, says Cuzick.

Related drugs called aromatase inhibitors also show promise – one has been shown to reduce the occurrence of breast cancer by 65 per cent (The New England Journal of Medicine, DOI: 10.1056/NEJMoa1103507).

Chemoprevention isn't just focusing on breast cancer. Last week, a study in The Lancet showed that aspirin dramatically reduces the risk of developing colorectal cancer in people with a family history of the disease. "We set out to see if aspirin would prevent cancer, and it does," says Burn, who led the study.

This is especially significant for developing countries, where cancer rates are escalating at a staggering rate (see "Poor countries need cancer drugs").

Burn and his colleagues studied 861 people with a hereditary form of colorectal cancer called Lynch syndrome, who began taking two 300-milligram tablets of aspirin a day or a placebo at some point between 1999 and 2005.

By 2010, there had been 19 new colorectal cancers in those who had taken aspirin and 34 in the placebo group.

In people who had taken aspirin for more than two years the effects were even more pronounced (The Lancet, DOI: 10.1016/S0140-6736(11)61049-0).

"It provides the first evidence that aspirin is effective in reducing the very high risk of cancer that these individuals have," says Peter Rothwell of the University of Oxford, who earlier this year found that a daily dose of 75 mg of aspirin for more than five years reduced the risk of dying from all cancers by 34 per cent.

Both Burn and Rothwell say they now regularly take aspirin for cancer prevention, but emphasise that self-medication is a personal decision: everyone has to weigh up the pros and cons for themselves.

 "Up until now, the main reason to take aspirin was to prevent vascular events.

I think it will become clear that cancer prevention is the main benefit of aspirin in healthy middle-aged people," says Rothwell.

Lung cancer is another disease where preventative therapy could reap rewards: especially for the millions of ex-smokers who remain at increased risk of disease.

 In a trial of 152 smokers and former smokers, a drug called iloprost significantly reduced abnormalities in cells lining the airways over the course of six months in those who had kicked the habit, but not in current smokers.

"If this holds up, it suggests that former smokers could reduce their risk of developing lung cancer by taking a drug," says Robert Keith of the University of Denver in Colorado, who also presented his results in Boston last week. Iloprost is a synthetic version of a naturally occurring molecule called prostacyclin, which can suppress cell growth and division.

Bringing such preventative drugs to market may not be so easy, however.

One of the biggest barriers is the need to test these drugs in large numbers of healthy individuals, which will inevitably produce side-effects in some people.

"Chemoprevention is tremendously appealing, but it is a more difficult path to traverse than developing a therapeutic drug," says Michael Thun of the American Cancer Society.

It is also an issue for people like Cuzick, who want tamoxifen and related drugs made available as a precaution for people at high risk. "Treatment can't be the whole answer," he says.

"We've got to do something about prevention."

IT MAY sound like a strange brew, but green tea and red light could provide a novel treatment for Alzheimer's disease.

Together, the two can destroy the rogue "plaques" that crowd the brains of people with the disease.

The light makes it easier for the green-tea extract to get to work on the plaques.

Andrei Sommer at the University of Ulm in Germany, and colleagues, have previously used red light with a wavelength of 670 nanometres to transport cancer drugs into cells.

The laser light pushes water out of the cells and when the laser is switched off, the cells "suck in" water and any other molecules, including drugs, from their surroundings.

Now, Sommer's team have found that the same technique can be used to destroy the beta-amyloid plaques in Alzheimer's.

These plaques consist of abnormally folded peptides, and are thought to disrupt communication between nerve cells, leading to loss of memory and other symptoms.

The team bathed brain cells containing beta-amyloid in epigallocatechin gallate (EGCG) - a green-tea extract known to have beta-amyloid inhibiting properties - at the same time as stimulating the cells with red light. Beta-amyloid in the cells reduced by around 60 per cent.

Shining the laser light alone onto cells reduced beta-amyloid by around 20 per cent (Photomedicine and Laser Surgery, DOI: 10.1089/pho.2011.3073).

It can be difficult getting drugs into the brain, but animal experiments show that the green-tea extract can penetrate the so-called blood-brain barrier when given orally together with red light.

The light, which can penetrate tissue and bone, stimulates cell mitochondria to kick-start a process that increases the barrier's permeability, says Sommer.

There is no reason why other drugs that attack beta-amyloid could not be delivered to the brain in the same way, he adds.

"This important research could form the basis of a potential treatment for Alzheimer's, with or without complementary drug treatment," says Mario Trelles, medical director of the Vilafortuny Medical Institute in Cambrils, Spain.

"The technique described could help to regulate and even stop the appearance of this disease," he adds.

This article has been edited since it was first posted.

For the first time, astronomers have found a planet-forming disc around a star that is awash with frozen water.

The discovery adds credence to the idea that Earth got its water from comets – especially as the disc seems to contain enough water to fill Earth's oceans thousands of times over.

Hot water vapour has previously been detected in the inner part of the planet-forming discs of nascent, alien solar systems.

But this is too close to the central star to be incorporated into the forming planets.

By contrast, the new observations are of water in the form of ice grains, which can exist only in the frigid outer reaches of a planet-forming disc.

It is there that they can ultimately coalesce into planets and comets.

Before planets form, the disc of material surrounding young stars is mostly gas.

Astronomers can probe the contents of that gas by analysing the spectra of the light it emits.

Earlier observations had found organic materials like carbon monoxide and cyanide in such discs, but because Earth's own atmosphere is so damp, it interferes with the detection of alien water from the ground.

UV puffs

To solve this problem, Michiel Hogerheijde of the Leiden Observatory in the Netherlands and colleagues used the Herschel space observatory to get above Earth's clouds and observe a young star called TW Hydrae, which has just over half the mass of the sun and is 175 light years away.

Earlier models suggested that there should be water in the outer frigid regions of TW Hydrae's disc, where it would be locked up in ice and therefore invisible to Hershel's infrared eyes.

But when those dust grains get hit by ultraviolet photons from their host star, they give off little puffs of water vapour that emit light in the wavelengths Herschel can see.

Hogerheijde and colleagues found a spectral signature that can be attributed to ice reservoirs about five one-thousandths of the mass of Earth's oceans.

But for every gram of vapour they spotted directly, there should be thousands of grams still frozen.

The team inferred that the total ice reservoir in the disc should amount to several thousands of Earth's hydrosphere.

Alien hope

The team further confirmed that the water they were seeing came from the frigid, comet-forming region of the disc via the signatures of two different types of water molecule, which form at different temperatures.

If the two hydrogen atoms in water have the same quantum spin, the water is called "ortho", and if they're different, it's called "para".

The ratio of these two versions of water in TW Hydrae's disc suggested that much of it formed cold – in other words in the frigid outer regions of the disc.

Some astronomers think Earth got its oceans from comets that smashed into the infant planet after it had cooled.

The discovery of cold water around TW Hydrae, combined with the recent discovery of a comet storm in a young planetary system suggests this scenario is a distinct possibility.

That could be good news for the prospect of life on dry exoplanets that are waiting for water.

 "If this is true in all systems, there is certainly a lot of water around," said Rachel Akeson of the NASA Exoplanet Science Institute at the California Institute of Technology, who was not involved in the new study.

 "It may increase the chance that life can develop on these planets."

Journal reference: Science, DOI: 10.1126/science.1208931

An earthquake of magnitude 7.2 struck Turkey yesterday, killing more than 200 people and injuring thousands.

Turkey is one of the most quake-prone countries in the world. Most of it lies on the Anatolian plate, a small wedge-shaped tectonic plate that is being squeezed westwards as the Arabian plate to the east slams into the Eurasian plate.

Many of Turkey's most severe quakes occur on one of the two faults that flank the Anatolian plate – the north and the east Anatolian faults. Between 1939 and 1999 Turkey's major earthquakes were marching westward along the north Anatolian fault, prompting fears that Istanbul – which lies near the fault – would eventually shake. In 1999 a magnitude-7.6 quake struck near Izmit, just 70 kilometres from Istanbul, killing around 17,000 people.

Since 2003, however, activity has shifted to the east Anatolian fault. In that year more than 100 people died after a quake near the city of Bingöl. The east fault slipped again last year, and the resulting 6.1-magnitude quake killed 51.

According to the US Geological Survey, yesterday's earthquake hit at 1.41 pm local time (1041 GMT) at a depth of 20 kilometres. Its epicentre was 16 kilometres north-east of Van in eastern Turkey, which places it near the junction of the two Anatolian faults. Here, tectonic activity is dominated by the Bitlis suture zone – a broad zone of compression caused by the collision of the Arabian and Eurasian plates.

"Since yesterday's quake is in the junction it's hard to know which fault was responsible," says Kevin McCue, director of the Australian Seismological Centre in Canberra. However, the USGS is now reporting that the style of tectonic activity is consistent with compressional activity within the Bitlis suture zone

Faster, safer modern cars may make higher speed limits appealing, but the human body is still stuck in the slow lane

Whenever a speed limit change is proposed there is a good deal of public debate.

The British government's recent call to allow drivers to do 80 miles per hour (129 kilometres per hour) on some of the country's fastest roads, instead of the current 70mph (113kph), is no different.

Should we raise the speed limit to take advantage of the greater capabilities of modern vehicles?

Would this increase casualties? How should limits be enforced?

These are just a few of the questions that provoke endless debate.

Speed is not the only factor in crashes – no one would argue otherwise.

However, its importance for public health is that it is easily experimented on.

Contrast that with driving while tired, which is less easy to measure and change.

There is no doubt that over the past 30 years vehicles have evolved to go faster with consummate ease.

Before taking full advantage of this we might consider whether we, the drivers, have evolved much over the same period.

Unfortunately our reaction times are not any faster, nor are our bodies any better at withstanding the forces involved in a crash.

A human who tries really hard can sprint at about 30kph.

To reach even that speed, we have to put lots of energy into the system, our heart is pumping, we have the wind in our face and massive experience of movement – in other words, we have overwhelming biological feedback to tell us just how fast we are going.

Velocity blindness

When we drive a car, however, the energy input is the small movement of a large toe.

The output, in contrast, is that we can easily travel at more than four times the maximum speed for which we have been designed – and with almost no experience of movement.

The feedback to the brain from the legs, heart and lungs when we are driving is effectively that of no movement.

Add to that the fact that prolonged exposure to speed reduces perceived velocity and that speed cues such as engine noise are systematically eliminated in modern vehicles, and it is no wonder there are some challenges in obeying limits.

We have the speedometer, of course, but this hardly provides visceral feedback – and, bizarrely, about half of the dial is devoted to illegal speeds.

It is rare for any other product to broadcast its illegal capabilities like this.

For years, the general message from governments has been that, for safety reasons, a reduction in speed is good because it reduces casualties.

But this has been difficult to get across.

Messages such as "at 35mph you are twice as likely to kill someone as at 30mph" may be hard to appreciate if you assume that energy increases linearly with speed – in fact, it rises with the square of the velocity.

The transfer of that energy to the human body is the problem.

The evidence on the relationship between speed and casualties is unambiguous whichever way it is examined.

For example, raising the 55mph (89kph) speed limit to 65mph (105kph) in the US was estimated to have increased fatalities by 15 per cent (American Journal of Public Health, vol 79, p 1392).

So what criteria should we use to define a limit? Two present themselves: functionality and survivability.

Survival speeds

Different types of road have different functions: access roads, which border residential and shopping areas; distribution roads, which need more entry and exit points; and through roads such as freeways and motorways which are for uninterrupted movement, with limited entry and exit.

Survivability refers to the body's capacity to tolerate the energy transfer in accidents.

Evidence shows that on access roads, where crashes involving pedestrians are likely, a 20mph (30kph) limit is appropriate.

On distribution roads, where side impacts are likely – when a car might ram into the side of another that is pulling out of a side road, for instance – the limit should be 30mph (50kph). In situations without pedestrians and where side impacts and head-on collisions are improbable – motorways and freeways – the limit should be 60 to 70mph (100 to 110kph).

Getting drivers to stick to limits, be they new or old ones, is another thing.

Deterrence is an obvious route.

Deterrence theory, derived from the work of the 18th-century judicial theorist Cesare Beccaria and the 19th-century philosopher and social reformer Jeremy Bentham, emphasises the certainty, severity and imminence of punishment.

The certainty of punishment has the clearest deterrent effect, which is problematic for speed enforcement because it relies on an uncertain police presence.

This can be solved by speed cameras, which have themselves stimulated a good deal of media debate.

Controversy has focused on whether their goal is safety or revenue generation.

Policy-makers can tackle this by emphasising casualty reduction: for instance, they can place cameras at accident locations, allocate fines to road safety, advertise the accident location by highly visible cameras and prior warning signs, and offer education for first-time offenders.

It may be important for politicians to distinguish between media debate and public concern on this issue. For example, Damian Poulter – a colleague at the University of Reading, UK – and I examined the UK government's British Crime Survey to determine what people are concerned about in their local communities.

In comparison with a range of antisocial behaviours such as race attack, drugs, intimidation and noisy neighbours, speeding was the top concern (Accident Analysis and Prevention, DOI: 10.1016/j.aap.2006.08.015).

So the challenge for governments who wish to change limits is complex.

This is particularly so for those wishing to raise them, which is a less familiar path.

What is clear, given the historical evidence and our biology, is that if they choose to permit faster driving, they must accept their part in the increased casualties that will follow.

Frank McKenna is a psychologist at the University of Reading, UK, and director of Perception and Performance, which provides consultancy on road safety to companies and government departments

What would you like in return for your organs?

As demand for organs continues to outstrip supply, ethical ways of rewarding people for donating need to be brought in, says a new report from the Nuffield Council on Bioethics, based in London.

Top of the list is the idea of paying for the funerals of people who agree that their organs can be transplanted after they die.

Almost 8000 people are waiting for a transplant in the UK.

Although 18 million people are signed up to the British organ donor register, three people every day die waiting for a donor.

The council hopes that offering to pay funeral expenses would encourage more people to sign up.

If introduced, it would be the first such scheme in the world. Keith Rigg, a consultant transplant surgeon and co-author of the report, says a pilot study should be introduced.

The average British funeral costs nearly £7000, but a funeral-for-transplant scheme would quickly pay for itself.

Kidney transplants alone save the taxpayer around £13000 a year per patient, as they remove the need for costly dialysis. Rigg says that reduced treatment after organ transplantation saved the British National Health Service more than £50 million in 2008.

Opt in or opt out?

In the UK, the donor must have explicitly given consent before their organs can be taken, but in some other countries, such as Spain, consent is assumed unless a person has registered their refusal.

There have long been calls for such an opt-out system to be introduced in the UK as a way to increase donor numbers.

Opt-out systems might seem a sure-fire way of getting more organs, but the report suggests that they are not. "There is uncertainty about whether or not an opt-out system could lead to more organs being donated," says Rigg.

Although Spain has the highest rate of organ donation in the world, Sweden, which also practises opt-out, has a rate lower than the UK.

There will always be uncertainty about what the person really wanted with an opt-out system, says Bobbie Farsides, a medical ethicist at Brighton and Sussex Medical School, UK.

The absence of a request to opt out could be a sign of the donor being confused or misinformed about the process, rather than a willingness to donate.

Living free

The report also made recommendations about live organ donation.

To prevent donors being exploited or harmed by repeat donations, it is against European law to offer payments directly for organs – and the authors say altruistic donation should remain the rule.

Any additional money given over expenses could undermine the motive of helping others, they say.

For egg and sperm donations, however, they recommend that a cap of £250 for expenses should be increased to match the lost earnings of anyone willing to donate.

With half of all fertility clinics in the UK short of sperm, and nearly all clinics short of eggs, no financial barrier should prevent donation, says Farsides.

Price of eggs

Women who are willing to suffer the discomforts and possible health effects of donating eggs for medical research should be paid over and above the expenses given to those donating for fertility treatment, however, the report concluded.

Participants in phase I clinical trials often receive hundreds of pounds for their involvement, says Marilyn Strathern, an anthropologist who also worked on the report.

"Donating eggs for research purposes is different from donating to help someone else's treatment.

You're not trying to help a particular individual – you are more a participant in a research exercise," she says.

A registry should be set up to ensure donations are monitored and regulated, she adds.

Women who are on the pill when they pick a mate end up with longer-lasting relationships than those who are not.

The downside?

 Less satisfaction in the sack, apparently.

Studies show that women on the pill are attracted to different men than when they are not on the pill.

One explanation is that because the pill simulates pregnancy, women tend to head for traits associated with fidelity.

Now the effects of these choices have been tested for the first time outside the lab.

Craig Roberts from the University of Stirling, UK, and colleagues surveyed 1000 women who had been taking the pill when they met their partner and 1500 who were not.

All of the women had at least one child with this partner.

The team measured the women's levels of sexual and general relationship satisfaction, as well as how long the relationship lasted.

Women on the pill when they met their partner were significantly less sexually satisfied, but reported higher levels of general satisfaction in the relationship, such as financial support and partner loyalty.

They were also more likely to stay together.

During fertile cycle phases, women seek out traits such as masculine faces, which are associated with good genes but also infidelity.

 Because women on the pill don't have such hormone shifts they may be more sensitive to traits which lead to longer relationships, says Roberts.

Too much alcohol dulls more than your wits.

It also weakens your immune system and could make you much more vulnerable to viruses, including HIV.

To see how alcohol affects resistance to infection, Gyongyi Szabo of the University of Massachusetts Medical School in Worcester and colleagues exposed monocytes – white blood cells involved in the front-line defence against infection – to chemicals that mimic viruses and bacteria.

 Half of the cells were also soused in the levels of alcohol that a person might have in their blood after quaffing four or five alcoholic drinks daily for a week.

Alcohol blunted the monocytes' defences.

 When the over-the-limit cells were exposed to a virus mimic, they produced only a quarter as much of the virus-fighting signalling molecule called type-1 interferon as teetotal monocytes made.

"Interferon is pivotal, the first response to any viral infection," says Szabo. "There's no viral elimination without it."

Lowered immunity

Monocytes exposed to a bacterial chemical suffered a double blow when inebriated.

Not only did they make half as much type-1 interferon as their abstemious equivalents, they also overproduced an inflammatory chemical called tumour necrosis factor-alpha.

Although important for initiating inflammatory responses to bacteria, continued production of this chemical can damage tissue.

Szabo says that the results fit with evidence from medical records that chronic heavy drinkers with HIV die sooner than non-drinkers.

They also fit with earlier studies showing that the immune system of heavy drinkers might be less vigilant against cancer.

Szabo says heavy drinkers should beware of damaging their immune systems.

Next, she hopes to see if alcohol makes flu vaccinations less effective.

Mark Hutchinson of the University of Adelaide in South Australia says that the results tally with post-mortem data showing that chronic drinkers have less immune chemicals in their blood than normal.

Brain and blood

In another study published this week, Hutchinson and his colleagues show in mice that the same monocytes, when situated in the brain, may play a part in drinkers' clumsiness.

"We're dealing with brain immune cells, which appear to respond to alcohol differently from blood immune cells," says Hutchinson.

His team discovered that blocking an antibacterial receptor on monocytes in the brain stopped mice being so clumsy when exposed to alcohol.

"It sure is a complicated story we've only scratched the surface of," Hutchinson says.

Journal references: immune system: BMC Immunology, in press; clumsiness: British Journal of Pharmacology, DOI: 10.1111/j.1476-5381.2011.01572.x

Editorial: "The good news about how food tweaks our genes"

CONSIDER the Brussels sprout: small, unassuming and ostensibly good for you.

This is no mere side dish.

A landmark study suggests that this dinky member of the cabbage family - along with rice, broccoli and possibly all the plants you eat - changes the behaviour of your genes in ways that are new to science.

In what is the strongest evidence yet that the genetic material in food survives digestion and circulates through the body, fragments of plant RNA have been found swimming in the bloodstreams of people and cows.

 What's more the study by Chen-Yu Zhang of Nanjing University in China and his colleagues shows that some of these plant RNAs muffle gene expression and raise cholesterol levels in mice.

The discovery opens up a new way to turn food into medicine: we may be able to design plants that change our genes for the better.

The genetic material in question is microRNA - tiny strands of RNA between 19 and 24 "letters" or nucleotides long. It is found in almost all cells with a nucleus and travels from cell to cell in the blood.

Zhang and his colleagues wondered whether all the miRNA strands in our blood are made by our cells - or whether some comes from our food instead.

To begin, the team drew blood from 31 healthy Chinese men and women, and also from cows.

They treated the samples with sodium periodate, an oxidising agent that modifies mammalian miRNA so that it cannot be sequenced.

Crucially, it leaves the plant versions untouched. Zhang found some 30 known plant miRNAs floating in the blood of the people and cows.

Two miRNAs were present in particularly high concentrations: MIR168a and MIR156a, which we will call 168a and 156a.

They are abundant in rice and members of the Brassicaceae family, including the Brussels sprout, broccoli, cabbage and cauliflower.

Surprisingly, Zhang found 168a and 156a in the livers, small intestines and lungs of mice.

Given the prominence of rice in the Chinese diet - coupled with the fact that cooking does not destroy the plant miRNAs - Zhang concluded that those in the human blood samples came from food.

That plant miRNA survives digestion and circulates through the body was surprising enough. But Zhang wanted to know whether plant miRNA remains functional in animal blood.

Like a genetic volume control, miRNA muffles or amplifies gene expression by binding to strands of messenger RNA and preventing enzymes from translating the strands into a protein.

To find out if 168a tweaks gene expression in animals, Zhang's team searched the human, rat and mice genomes for sequences that complemented 168a.

They found around 50 genes that 168a might turn up or down, including the gene for LDLRAP1, a liver protein that removes "bad cholesterol" from the blood.

In a series of experiments, Zhang and the team found that not only does 168a survive in animal cells, it can also change gene expression.

First, Zhang added 168a to a dish of human intestinal cells. The cells packaged the 168a into tiny bubbles and released them.

Zhang poured these bubbles onto mammalian liver cells, which soon began producing unusually low levels of LDLRAP1.

Then Zhang fed mice raw rice or injected them with 168a, and found that levels of this protein dropped and levels of cholesterol rose.

When he injected the mice with a genetic sequence designed to inactivate 168a, levels of the cholesterol-removing protein did not drop.

Together, the evidence suggests that, in mice at least, 168a from rice survives digestion, inhibits production of a protein and boosts cholesterol levels in the blood.

Put simply, a plant miRNA is capable of raising cholesterol levels in mice (Cell Research, DOI: 10.1038/cr.2011.158).

Zhang is unsure how the miRNAs escape unscathed from the caustic soup of digestive fluids and enzymes in the gut.

But substantial research suggests that not all genetic matter from food dissolves in the stomach and intestines.

For instance, an essential photosynthesis gene found in soya bean leaves turned up in the intestines, liver and spleen of mice fed the leaves.

And it was recently revealed that the hypnotically green sea slug Elysia chlorotica

steals genes for photosynthesis from the algae it eats.

 Researchers also discovered that the bacteria in Japanese people's guts have sponged up genes from ocean bacteria that linger on seaweed.

Even if RNA or DNA does not pass unscathed from food to eater, food can change gene expression in other ways.

For example, cosmetics researchers recently suggested that a pill containing a mix of food extracts can influence our genes and boost collagen production in the skin, reducing the appearance of wrinkles (New Scientist, 24 September, p 10).

If Zhang's findings are replicated, we may discover that our blood is swimming with RNA from all kinds of plants.

To date, all investigation of this possibility has been motivated by concerns that genes from genetically modified crops could harm health (see "Let's talk about GM crops"). But the new study opens the possibility of designing diets and plants with therapeutic effects.

"You can bet this will create an absolute flurry of research activity" as scientists race to discover how genetic information in our food changes our health, says Ed Stellwag of East Carolina University in Greenville.

Peter Waterhouse of the University of Sydney, Australia, sees the potential for engineering medicinal plants but adds that for now this remains unchartered territory - mostly. Zhang is investigating whether miRNAs in a Chinese herb can knock out the influenza virus, but remains tight-lipped about the results.

"This will expand our idea of nutrients by including miRNA as functional component of food," says Moon-Suhn Ryu at the University of Florida in Gainesville.

"This is going to introduce a new field of research, especially in nutritional science - it's such a novel concept."

SLOWLY and almost imperceptibly the seas are rising, swollen by melting ice and the expansion of seawater as it warms. But there's another source of water adding to the rise: humanity's habit of pumping water from underground aquifers to the surface. Most of this water ends up in the sea.

Not many scientists even consider the effects of groundwater on sea level, says Leonard Konikow of the United States Geological Survey in Reston, Virginia. Estimates were published as far back as 1994 (Nature, DOI: 10.1038/367054a0), but without good evidence to back them up, he says. The last report of the Intergovernmental Panel on Climate Change said that changes to groundwater reserves "cannot be estimated with much confidence".

Konikow measured how much water had ended up in the oceans by looking at changes in groundwater levels in 46 well-studied aquifers, which he then extrapolated to the rest of the world. He estimates that about 4500 cubic kilometres of water was extracted from aquifers between 1900 and 2008.

That amounts to 1.26 centimetres of the overall rise in sea levels of 17 cm in the same period (Geophysical Research Letters, DOI: 10.1029/2011gl048604).

That 1.26 cm may not seem like much, but groundwater depletion has accelerated massively since 1950, particularly in the past decade. Over 1300 cubic kilometres of the groundwater was extracted between 2000 and 2008, producing 0.36 cm of the total 2.79-cm rise in that time. "I was surprised that the depletion has accelerated so much," Konikow says.

It's not clear if the acceleration will continue. Konikow points out that some developed countries are cutting back on aquifer use and even trying to refill them when there is plenty of rainfall. "I would like to see that implemented more," he says.

"While there remain significant uncertainties, Konikow's estimate is probably the best there is for groundwater depletion," says John Church of CSIRO Marine and Atmospheric Research in Hobart, Tasmania, Australia.

Konikow, Church and colleagues have used the data to compare the contributions of the different sources of sea-level rise and found that aquifer depletion is almost as significant as the ice melt in Greenland and Antarctica combined (Geophysical Research Letters, DOI: 10.1029/2011gl048794).

SOME people don't seem to believe anything they're told about flu. You'll often hear that the swine flu pandemic of 2009, along with the spectre of H5N1 bird flu, were "scares" backed by some conspiracy or other.

Of course, the 2009 pandemic was real, it just wasn't as bad as it could have been. Bird flu is about as bad as flu can get, and the only thing that has kept it at bay has been its inability to spread easily between people.

That may have been a temporary situation. Work reported last week suggests that just a few mutations could make H5N1 highly contagious in humans without losing its ability to kill 60 per cent of those it infects (see "Five easy mutations to make bird flu a lethal pandemic").

Now more than ever, the world needs its flu defences to be in order. The 2009 pandemic showed they aren't, with vaccine arriving late, in relatively few countries. And because the pandemic was limited, investment to improve vaccines is far from booming.

It should be. If vaccines are not ready fast enough after the next pandemic hits, there will still be conspiracy theories, but the "scare" will be all too real.

New Scientist tracks the evolution of our brain from its origin in ancient seas to its dramatic expansion in one ape – and asks why it is now shrinking

IT IS 30,000 years ago. A man enters a narrow cave in what is now the south of France. By the flickering light of a tallow lamp, he eases his way through to the furthest chamber. On one of the stone overhangs, he sketches in charcoal a picture of the head of a bison looming above a woman's naked body.

In 1933, Pablo Picasso creates a strikingly similar image, called Minotaur Assaulting Girl.

That two artists, separated by 30 millennia, should produce such similar work seems astonishing. But perhaps we shouldn't be too surprised. Anatomically at least, our brains differ little from those of the people who painted the walls of the Chauvet cave all those years ago. Their art, part of the "creative explosion" of that time, is further evidence that they had brains just like ours.

How did we acquire our beautiful brains? How did the savage struggle for survival produce such an extraordinary object? This is a difficult question to answer, not least because brains do not fossilise. Thanks to the latest technologies, though, we can now trace the brain's evolution in unprecedented detail, from a time before the very first nerve cells right up to the age of cave art and cubism.

The story of the brain begins in the ancient oceans, long before the first animals appeared. The single-celled organisms that swam or crawled in them may not have had brains, but they did have sophisticated ways of sensing and responding to their environment. "These mechanisms are maintained right through to the evolution of mammals," says Seth Grant at the Wellcome Trust Sanger Institute in Cambridge, UK. "That's a very deep ancestry."

The evolution of multicellular animals depended on cells being able to sense and respond to other cells - to work together. Sponges, for example, filter food from the water they pump through the channels in their bodies. They can slowly inflate and constrict these channels to expel any sediment and prevent them clogging up. These movements are triggered when cells detect chemical messengers like glutamate or GABA, pumped out by other cells in the sponge. These chemicals play a similar role in our brains today (Journal of Experimental Biology, vol 213, p 2310).

Releasing chemicals into the water is a very slow way of communicating with distant cells - it can take a good few minutes for a demosponge to inflate and close its channels. Glass sponges have a faster way: they shoot an electrical pulse across their body that makes all the flagellae that pump water through their bodies stop within a matter of seconds (Nature, vol 387, p 29).

This is possible because all living cells generate an electrical potential across their membranes by pumping out ions. Opening up channels that let ions flow freely across the membrane produces sudden changes in this potential. If nearby ion channels also open up in response, a kind of Mexican wave can travel along a cell's surface at speeds of several metres a second. Since the cells in glass sponges are fused together, these impulses can travel across their entire bodies.

Deep roots

Recent studies have shown that many of the components needed to transmit electrical signals, and to release and detect chemical signals, are found in single-celled organisms known as choanoflagellates. That is significant because ancient choanoflagellates are thought to have given rise to animals around 850 million years ago.

So almost from the start, the cells within early animals had the potential to communicate with each other using electrical pulses and chemical signals. From there, it was not a big leap for some cells to become specialised for carrying messages.

These nerve cells evolved long, wire-like extensions - axons - for carrying electrical signals over long distances. They still pass signals on to other cells by releasing chemicals such as glutamate, but they do so where they meet them, at synapses. That means the chemicals only have to diffuse across a tiny gap, greatly speeding things up. And so, very early on, the nervous system was born.

The first neurons were probably connected in a diffuse network across the body (see diagram). This kind of structure, known as a nerve net, can still be seen in the quivering bodies of jellyfish and sea anemones.

But in other animals, groups of neurons began to appear - a central nervous system. This allowed information to be processed rather than merely relayed, enabling animals to move and respond to the environment in ever more sophisticated ways. The most specialised groups of neurons - the first brain-like structure - developed near the mouth and primitive eyes.

Our view of this momentous event is hazy. According to many biologists, it happened in a worm-like creature known as the urbilaterian (see diagram), the ancestor of most living animals including vertebrates, molluscs and insects. Strangely, though, some of its descendants, such as the acorn worm, lack this neuronal hub.

It is possible the urbilaterian never had a brain, and that it later evolved many times independently. Or it could be that the ancestors of the acorn worm had a primitive brain and lost it - which suggests the costs of building brains sometimes outweigh the benefits.

Either way, a central, brain-like structure was present in the ancestors of the vertebrates. These primitive, fish-like creatures probably resembled the living lancelet, a jawless filter-feeder. The brain of the lancelet barely stands out from the rest of the spinal cord, but specialised regions are apparent: the hindbrain controls its swimming movements, for instance, while the forebrain is involved in vision. "They are to vertebrates what a small country church is to Notre Dame cathedral - the basic architecture is there though they lack a lot of the complexity," says Linda Holland at the University of California, San Diego.

Some of these fish-like filter feeders took to attaching themselves to rocks. The swimming larvae of sea squirts have a simple brain but once they settle down on a rock it degenerates and is absorbed into the body.

We would not be here, of course, if our ancestors had not kept swimming. And around 500 million years ago, things went wrong when one of them was reproducing, resulting in its entire genome getting duplicated. In fact, this happened not just once but twice.

These accidents paved the way for the evolution of more complex brains by providing plenty of spare genes that could evolve in different directions and take on new roles. "It's like the time your parents bought you the biggest Lego kit - with loads of different components to use in different combinations," says Grant. Among many other things, it enabled different brain regions to express different types of neurotransmitter, which in turn allowed more innovative behaviours to emerge.

As early fish struggled to find food and mates, and dodge predators, many of the core structures still found in our brains evolved: the optic tectum, involved in tracking moving objects with the eyes; the amygdala, which helps us to respond to fearful situations; parts of the limbic system, which gives us our feelings of reward and helps to lay down memories; and the basal ganglia, which control patterns of movements (see diagram).

Brainy mammals

By 360 million years ago, our ancestors had colonised the land, eventually giving rise to the first mammals about 200 million years ago. These creatures already had a small neocortex - extra layers of neural tissue on the surface of the brain responsible for the complexity and flexibility of mammalian behaviour. How and when did this crucial region evolve? That remains a mystery. Living amphibians and reptiles do not have a direct equivalent, and since their brains do not fill their entire skull cavity, fossils tell us little about the brains of our amphibian and reptilian ancestors.

What is clear is that the brain size of mammals increased relative to their bodies as they struggled to contend with the dinosaurs. By this point, the brain filled the skull, leaving impressions that provide tell-tale signs of the changes leading to this neural expansion.

Timothy Rowe at the University of Texas at Austin recently used CT scans to look at the brain cavities of fossils of two early mammal-like animals, Morganucodon and Hadrocodium, both tiny, shrew-like creatures that fed on insects. This kind of study has only recently become feasible. "You could hold these fossils in your hands and know that they have answers about the evolution of the brain, but there was no way to get inside them non-destructively," he says. "It's only now that we can get inside their heads."

Rowe's scans revealed that the first big increases in size were in the olfactory bulb, suggesting mammals came to rely heavily on their noses to sniff out food. There were also big increases in the regions of the neocortex that map tactile sensations - probably the ruffling of hair in particular - which suggests the sense of touch was vital too (Science, vol 332, p 955). The findings fit in beautifully with the widely held idea that early mammals were nocturnal, hiding during the day and scurrying around in the undergrowth at night when there were fewer hungry dinosaurs running around.

After the dinosaurs were wiped out, about 65 million years ago, some of the mammals that survived took to the trees - the ancestors of the primates. Good eyesight helped them chase insects around trees, which led to an expansion of the visual part of the neocortex. The biggest mental challenge, however, may have been keeping track of their social lives.

If modern primates are anything to go by, their ancestors likely lived in groups. Mastering the social niceties of group living requires a lot of brain power. Robin Dunbar at the University of Oxford thinks this might explain the enormous expansion of the frontal regions of the primate neocortex, particularly in the apes. "You need more computing power to handle those relationships," he says. Dunbar has shown there is a strong relationship between the size of primate groups, the frequency of their interactions with one another and the size of the frontal neocortex in various species.

Besides increasing in size, these frontal regions also became better connected, both within themselves, and to other parts of the brain that deal with sensory input and motor control. Such changes can even be seen in the individual neurons within these regions, which have evolved more input and output points.

All of which equipped the later primates with an extraordinary ability to integrate and process the information reaching their bodies, and then control their actions based on this kind of deliberative reasoning. Besides increasing their overall intelligence, this eventually leads to some kind of abstract thought: the more the brain processes incoming information, the more it starts to identify and search for overarching patterns that are a step away from the concrete, physical objects in front of the eyes.

Which brings us neatly to an ape that lived about 14 million years ago in Africa. It was a very smart ape but the brains of most of its descendants - orang-utans, gorillas and chimpanzees - do not appear to have changed greatly compared with the branch of its family that led to us. What made us different?

It used to be thought that moving out of the forests and taking to walking on two legs lead to the expansion of our brains. Fossil discoveries, however, show that millions of years after early hominids became bipedal, they still had small brains.

We can only speculate about why their brains began to grow bigger around 2.5 million years ago, but it is possible that serendipity played a part. In other primates, the "bite" muscle exerts a strong force across the whole of the skull, constraining its growth. In our forebears, this muscle was weakened by a single mutation, perhaps opening the way for the skull to expand. This mutation occurred around the same time as the first hominids with weaker jaws and bigger skulls and brains appeared (Nature, vol 428, p 415).

Once we got smart enough to innovate and adopt smarter lifestyles, a positive feedback effect may have kicked in, leading to further brain expansion. "If you want a big brain, you've got to feed it," points out Todd Preuss of Emory University in Atlanta, Georgia.

He thinks the development of tools to kill and butcher animals around 2 million years ago would have been essential for the expansion of the human brain, since meat is such a rich source of nutrients. A richer diet, in turn, would have opened the door to further brain growth.

Primatologist Richard Wrangham at Harvard University thinks that fire played a similar role by allowing us to get more nutrients from our food. Eating cooked food led to the shrinking of our guts, he suggests. Since gut tissue is expensive to grow and maintain, this loss would have freed up precious resources, again favouring further brain growth.

Mathematical models by Luke Rendell and colleagues at the University of St Andrews in the UK not only back the idea that cultural and genetic evolution can feed off each other, they suggest this can produce extremely strong selection pressures that lead to "runaway" evolution of certain traits. This type of feedback might have played a big role in our language skills. Once early humans started speaking, there would be strong selection for mutations that improved this ability, such as the famous FOXP2 gene, which enables the basal ganglia and the cerebellum to lay down the complex motor memories necessary for complex speech.

The overall picture is one of a virtuous cycle involving our diet, culture, technology, social relationships and genes. It led to the modern human brain coming into existence in Africa by about 200,000 years ago.

Evolution never stops, though. According to one recent study, the visual cortex has grown larger in people who migrated from Africa to northern latitudes, perhaps to help make up for the dimmer light up there (Biology Letters, DOI: 10.1098/rsbl.2011.0570).

Downhill from here

So why didn't our brains get ever bigger? It may be because we reached a point at which the advantages of bigger brains started to be outweighed by the dangers of giving birth to children with big heads. Or it might have been a case of diminishing returns.

Our brains are pretty hungry, burning 20 per cent of our food at a rate of about 15 watts, and any further improvements would be increasingly demanding. Simon Laughlin at the University of Cambridge compares the brain to a sports car, which burns ever more fuel the faster it goes.

One way to speed up our brain, for instance, would be to evolve neurons that can fire more times per second. But to support a 10-fold increase in the "clock speed" of our neurons, our brain would need to burn energy at the same rate as Usain Bolt's legs during a 100-metre sprint. The 10,000-calorie-a-day diet of Olympic swimmer Michael Phelps would pale in comparison.

Not only did the growth in the size of our brains cease around 200,000 years ago, in the past 10,000 to 15,000 years the average size of the human brain compared with our body has shrunk by 3 or 4 per cent. Some see this as no cause for concern. Size, after all, isn't everything, and it's perfectly possible that the brain has simply evolved to make better use of less grey and white matter. That would seem to fit with some genetic studies, which suggest that our brain's wiring is more efficient now than it was in the past.

Others, however, think this shrinkage is a sign of a slight decline in our general mental abilities. David Geary at the University of Missouri-Columbia, for one, believes that once complex societies developed, the less intelligent could survive on the backs of their smarter peers, whereas in the past, they would have died - or at least failed to find a mate.

This decline may well be continuing. Many studies have found that the more intelligent people are, the fewer children they tend to have. More than ever before, intellectual and economic success are not linked with having a bigger family. If it were, says Rendell, "Bill Gates would have 500 children."

This evolutionary effect would result in a decline of about 0.8 IQ points per generation in the US if you exclude the effects of immigration, a 2010 study concluded (Intelligence, vol 38, p 220). However, nurture matters as well as nature: even if this genetic effect is real, it has been more than compensated for by improved healthcare and education, which led a steady rise in IQ during most of the 20th century.

Crystal-ball gazing is always a risky business, and we have no way of knowing the challenges that humanity will face over the next millennia. But if they change at all, it appears likely that our brains are going keep "devolving" - unless, of course, we step in and take charge.

David Robson is a features editor at New Scientist

Concussion has long been seen as a temporary and fairly harmless affliction.

But the repercussions can last a lifetime

THE two men in the hospital ward had both hit their heads in car accidents, but that was where the similarities ended.

One would spend weeks unconscious in critical care, near to death.

The other had only a mild concussion; he never lost consciousness, but somehow didn't feel quite right.

Yet months later their roles were reversed.

"The one with the severe injury is almost back to normal function," says Douglas Smith, director of the Center for Brain Injury Repair at the University of Pennsylvania in Philadelphia, "and the one with concussion can't go back to work, probably ever."

Smith's two patients illustrate one of the frustrating paradoxes of head injuries: even seemingly mild impacts can have devastating long-term consequences.

And we have no way of predicting who will fully recover and who will have lingering problems.

Concussion, or mild traumatic brain injury as doctors call it, has long been seen as a benign and temporary affliction.

But over the past decade there has been growing realisation that longer-term symptoms can affect between 10 and 15 per cent of those diagnosed with it.

These range from fuzzy thinking and memory lapses to, for the most unfortunate, serious neurological conditions such as premature Alzheimer's disease.

In fact, concussion is thought to be the single biggest environmental cause of Alzheimer's.

Even a mild impact can double the risk of developing early dementia, according to a massive study of older military veterans in the US, which was presented at the Alzheimer's Association International Conference in July (bit.ly/pDhlHJ). In that, the risk jumped from 7 to 15 per cent.

It now appears that concussion is not a single, discrete event, but can be the beginning of a degenerative disease lasting weeks, months or longer.

"It is an injury that can keep on taking," Smith says.

As details of that process emerge, researchers are beginning to identify where its progression might be blocked.

And the advent of new brain scanning techniques may some day let us identify people in the emergency room who are most at risk of long-term problems.

If the research lives up to its promise, that would be good news for us all.

The most common causes of concussion are falls and car or bike accidents, and we face an estimated 1 in 4 chance of sustaining a concussion in our lifetime.

That's not the only reason research into concussion is getting record levels of funding.

On the battlefield, mild traumatic brain injury has been dubbed the signature injury of service in Iraq and Afghanistan, affecting up to 40 per cent of soldiers who see combat duty.

One of the biggest threats to soldiers stationed in these countries is from roadside bombs, often improvised from cast-off artillery shells or other weapons.

The introduction of better body and vehicle armour means more soldiers than ever are surviving such blasts, but are left with concussion and other serious injuries.

Then there are the hundreds of thousands of people with sports-related concussions each year, most of whom never go to hospital.

This group is especially worrying, since many players of contact sports like rugby, American football or ice hockey experience repeated concussions, and evidence is growing that these serial injuries can heighten the risk of severe brain degeneration years or even decades later.

The risk for athletes is getting increasing attention, particularly in the US, where 75 retired American footballers recently sued the National Football League for doing too little to protect players from the effects of repeated concussions.

And the media has jumped on cases such as that of former American footballer David Duerson, who committed suicide last year fearing dementia was setting in.

Duerson was careful to shoot himself in the chest, not the head, so that scientists at the Boston-based Center for the Study of Traumatic Encephalopathy could examine his brain.

Sure enough, Duerson's autopsy revealed telltale signs of brain deterioration, though it has not yet been proven that these changes caused his symptoms.

One reason it has taken so long to understand concussion is that a blow to the head can cause many bad things to happen to the brain, including bleeding, swelling and the death of nerve cells.

With all this chaos going on, it has been difficult to unpick which problems are the most important.

Nor is there a standard way to image the brain to gauge the severity of concussion.

"If you fall on your leg and it hurts, we can take an X-ray, and it really helps us decide what to do," says Jeffrey Bazarian, a neuroscientist at the University of Rochester Medical Center in New York state. "We don't have something like that for concussion."

Leaky neurons

For a long time we have suspected that the fibres, or axons, of nerve cells may be particularly vulnerable to concussive blows.

"A neuron is like a big, long spaghetti strand with the cell body at one end.

This makes it very susceptible to stretch," says Bazarian.

"When the head gets struck, the head rotates on the neck, and the brain rotates inside the skull. All those spaghetti strands get stretched."

Back in the 1980s, researchers thought that axon damage occurred only after the most severe impacts, because it was seen in the brains of people who had died from their injuries.

Gradually animal studies began to point to damage occurring from milder blows too, although direct evidence about what happens in humans was still lacking.

That is now changing thanks to more sensitive forms of brain scanning.

One in particular, known as diffusion tensor imaging (DTI), is especially valuable.

Developed only within the past few years, this technique tracks the movements of water molecules in the brain.

Normally they move predominantly along the length of the axons.

After a blow to the head, however, water also moves laterally, a sign that stretching has made the axons leakier.

The worst of this stretch damage occurs to axons in the brain's frontal lobes, furthest from the axis of rotation at the neck, according to unpublished DTI studies led by Jamshid Ghajar, a neurosurgeon and president of the Brain Trauma Foundation in New York City.

This could explain why people with concussion experience problems concentrating and planning actions - key functions of the frontal lobes - rather than, say, loss of vision or movement control, which are handled in other parts of the brain.

This idea also explains why animals are less susceptible to concussion than people: animal brains are too small to experience the same forces.

"If you're in the front seat of your car with your dog next to you, and you're in an accident, you might lose consciousness for 5 minutes," says Smith.

"When you wake up, your dog is licking you.

That's because his 60-gram brain doesn't experience the deformation that your 1500-gram brain does."

The immediate effects of axon stretching are fairly well understood.

The axon surface is studded with molecular channels designed to transport sodium or calcium ions into the axon when the neuron fires an electrical signal.

When the axon is stretched, these ion channels are pulled open and the ions flow in, sparking a storm of electrical activity.

In response, nerve cells frantically pump the ions back out again.

This rapidly depletes the brain's store of energy - especially since the calcium influx chokes off the mitochondria, the cells' energy source.

The result is that the brain suddenly flips from overactivity to a state of exhaustion that can last for several days.

This exhaustion may help to explain why athletes are prone to repeat concussions shortly after returning to play.

Indeed, more than three-quarters of athletes who experience repeat concussions in a single season incur the second within 10 days of the first, according to a study by Michael McCrea, director of brain injury research at the Medical College of Wisconsin in Milwaukee (Journal of the American Medical Association, vol 290, p 2549). And the second injury is often worse, McCrea adds.

"If you pile a second injury on top of the first one, before the brain has fully recovered, that's not good."

If this short-lived leakiness and exhaustion were all that happened after a concussion, people should recover completely once its effects had passed - and that is what happens in most cases.

But for some it is the beginning of a longer, more serious disease process in which nerve cells continue to degenerate. In unpublished work, Ghajar used DTI to track the health of people who had experienced mild brain injuries.

He found that axon damage was initially confined to the frontal lobe.

But in those who still had symptoms one year later, the damage had spread to other regions of the brain that were not affected initially.

"I think it's a domino effect," says Ghajar. "The question is: why?"

Answers are now beginning to emerge.

 For one thing, the initial influx of calcium ions leads to local inflammation.

Also, even mild injuries can cause long-term problems with the sodium channels.

Using isolated rat axons stretched by a puff of air, Smith's team has shown that the cells respond to a single gentle stretch by adding more sodium channels within hours of the injury.

"What we suspect is that although the stretch injury was mild, it still disrupted the sodium channels enough that more were needed," says Smith.

The extra channels may make it easier to restore normal ion flow after the injury, but they also make the cells leakier after a second injury (Journal of Neuroscience Research, vol 87, p 3620).

Stretching can also damage an axon's internal structure, especially the microtubules that transport molecules around the cell.

The microtubules behave like Silly Putty, says Smith - when pulled gradually they stretch smoothly, but when jerked suddenly they become brittle and can snap.

If that happens, it's like breaking a train track: all the microtubules' cargo derails at the break.

When this happens in an axon, which it seems to after a blow to the head, the long-term consequences can be dire.

That's because one of the main cargoes within axons is a molecule called amyloid precursor protein, which helps to regulate connections betwen nerve cells.

When APP piles up at the site of a microtubule break, it is broken down into a smaller protein, amyloid beta.

This can accumulate within the axons and has also been found aggregated into fibrous plaques within the brain, presumably after being released from dying cells.

These amyloid plaques are all too well known - they are a hallmark of Alzheimer's and other degenerative brain diseases. In Alzheimer's, amyloid beta causes changes to another protein known as tau, which also forms tangled plaques.

Tau tangles also turn up in people who have had repeated concussions, notes Mark Burns, a neuroscientist at Georgetown University in Washington, DC.

The usual suspects

The sequence of events in Alzheimer's is not exactly the same as what happens after head injury. In particular, the amyloid plaques appear to be temporary, not permanent, after head injury, and the tau plaques tend to occur deeper within the brain in Alzheimer's than after a head injury, says Burns.

But the fact that the same two proteins are suspects in both sorts of brain disease suggests a new lead to follow.

Indeed, Burns has more direct evidence linking amyloid beta to brain injury.

His team delivered mild-to-moderate brain injuries to mice, then tested their ability to walk along a 6-millimetre-wide beam. Uninjured mice can do it relatively easily, but injured ones can no longer muster the necessary coordination.

"They'll go from about five mistakes per 50 steps to 50 mistakes," says Burns.

Three weeks after injury, the mice have regained a little of their coordination, but still average 40 to 45 mistakes per 50 steps.

But when Burns gave mice a drug that blocks the production of amyloid beta, they recovered far more quickly, making just 25 mistakes per 50 steps after three weeks (Nature Medicine, vol 15, p 377).

A different drug - one that helps the body clear away unwanted amyloid beta instead of preventing its formation - also speeded up the mice's recovery, Burns's team reported earlier this year (Journal of Neurotrauma, vol 28, p 225).

All this suggests that amyloid beta and perhaps tau are involved in the dementia that sometimes follows concussion.

If so, then drugs to block the amyloid beta cascade may speed up recovery from concussion and ward off at least some of its long-term effects in people too.

 Much testing remains before this could happen, of course, but if they prove safe enough, such drugs could be given routinely after a concussion as a preventative measure.

Burns is not the only researcher with promising potential therapies to treat axon injury.

Earlier this year Ramona Hicks, who directs the brain injury repair programme at the US National Institute of Neurological Disorders and Stroke in Bethesda, Maryland, convened a workshop to review axonal injury and its treatment.

Though no therapies are yet ready for clinical use, Hicks is optimistic that our growing understanding of brain injury, and our increasing ability to peer into the injured brain with advanced imaging technologies like DTI, should yield much progress in the next few years.

If so, the day may come when physicians will know in the emergency room which concussed patients are at greatest risk of complications and begin treatment to prevent them.

That way athletes, blast-shocked soldiers and those who are simply unlucky enough to fall in the street will have a greater chance of complete recovery.

Bob Holmes is a consultant for New Scientist based in Edmonton, Canada

Smells shape our moods, behaviour and decisions, so why do they barely register in our conscious lives?

I TRY to forget about potential onlookers as I crawl around a central London park, blindfolded and on all fours.

With a bit of luck, the casual passer-by might not notice the blindfold and think I'm just looking for a contact lens.

In fact, I'm putting my sense of smell to the test, and attempting to emulate the sensory skills of a sniffer dog.

Just as a beagle can swiftly hunt down a pheasant using only its nasal organ, I am using mine to follow a 10-metre trail of cinnamon oil.

Such a challenge might sound doomed to failure.

After all, dog noses are renowned for their sensitivity to smells, while human noses are poor by comparison.

Yet that might be a misconception.

According to a spate of recent studies, our noses are in fact exquisitely sensitive instruments that guide our everyday life to a surprising extent.

Subtle smells can change your mood, behaviour and the choices you make, often without you even realising it.

Our own scents, meanwhile, flag up emotional states such as fear or sadness to those around us.

The big mystery is why we aren't aware of our nasal activity for more of the time.

Noses have certainly never been at the forefront of sensory research, and were pushed aside until recently in favour of the seemingly more vital senses of vision and hearing.

"There has been a lot of prejudice that people are not that influenced by olfactory stimuli, especially compared to other mammals," says Lilianne Mujica-Parodi, who studies the neurobiology of human stress at Stony Brook University in New York.

One of the first people to assert the relative unimportance of human smelling was Pierre Paul Broca, an influential 19th-century anatomist.

After comparing the proportion of the brain devoted to smell in different animals, he suggested that mammals can be classed into two broad groups: macrosmatic mammals, such as dogs, have a finely tuned sense of smell which they rely on to perceive the world, while we, along with other primates and the marine mammals, are microsmatic - we have a small and functionally redundant olfactory apparatus.

That idea seemed to fit with more recent studies in genetics, which found that the majority of mammals have genes coding for about 1000 different types of smell receptor.

 Most of these genes aren't expressed in humans, giving our noses just 400 different types of receptor (see chart).

Yet these findings may have been misleading.

 Brain scans now show that more of the brain is devoted to smell processing than Broca's anatomical studies would have suggested. And although we may have fewer types of receptor than other mammals, Charles Greer at Yale University has shown that the human nose and brain are unusually well connected, with each group of receptors linking to many more neural regions than is the case in other animals.

That should give us a good ability to process incoming scents.

Once researchers began looking, they found the nose to be far more sensitive than its reputation suggested.

One study, for example, found that we can detect certain chemicals diluted in water to less than one part per billion.

That means that a person can detect just a few drops of a strong odorant like ethyl mercaptan in an Olympic-sized pool.

We are also exceptionally gifted at telling smells apart, even distinguishing between two molecules whose only difference is that their structures are mirror images of one another (Chemical Senses, vol 24, p 161).

"That is fantastic sensitivity," says George Dodd, a perfumer and researcher at the olfaction group of the University of Warwick, UK.

What's more, it is becoming clear that the brain's olfactory centres are intimately linked to its limbic system, which is involved in emotion, fear and memory.

That suggests a link between smell and the way we think.

The power of smell will be no news to estate agents, who often advocate the smell of baking bread or brewing coffee to promote the sale of a house.

But there are more subtle and surprising effects too.

For instance, when Hendrick Schifferstein from Delft University of Technology and colleagues pumped the smell of orange, seawater or peppermint into a nightclub, the revellers partied harder - they danced more, rated their night as more enjoyable, and even thought the music was better - than when there was no added scent (Chemosensory Perception, vol 4, p 55).

Rob Holland and colleagues at the University of Utrecht in the Netherlands, meanwhile, have found that the hint of aroma wafting out of a hidden bucket of citrus-scented cleaner was enough to persuade students to clean up after themselves - even though the vast majority of them hadn't actually registered the smell (Psychological Science, vol 16, p 689).

Other work has found that scent can influence our cognitive skills. A study this year by William Overman and colleagues at the University of North Carolina Wilmington found that when men were subjected to a novel smell - either good or bad - during a gambling task used to test decision-making skills, they performed significantly worse than normal.

The researchers conclude the scent stimulated brain areas connected with emotion, making their decisions emotional rather than rational (Behavioral Brain Research, vol 218, p 64).

Smells also seem to direct our visual attention, and they may play a key role in consolidating memories too (see "Blast from the past").

Our sense of smell may even help us to pick up on the emotional state of those around us.

This idea has been highly controversial, but work by Mujica-Parodi suggests we can sense another's fear from their sweat.

At the time, she was working on a way to assess a person's vulnerability to stress, and needed a reliable way to scare her subjects, without socially loaded words or pictures that might mean different things to different people.

That's hard to do, says Mujica-Parodi: "You can't mug somebody in a scanner."

Freak out

The answer came from nature.

Rats are known to be able to smell each other's fear, leading them to "freak out" if placed in an empty cage in which another rat has just seen a predator. Mujica-Parodi figured humans might do the same thing.

To test the idea, her team took sweat samples from people doing a skydive for the first time.

When they presented the samples to unrelated subjects in an fMRI scanner, they saw activation of the amygdala - the brain area that normally lights up in studies of emotion.

This did not happen when sweat samples came from the same skydivers pounding a treadmill.

Mujica-Parodi's team next tested whether the smell of fear sweat affected people's responses to various facial expressions - angry, ambiguous or neutral.

Normally, we would pay more attention to angry faces, because they pose a threat, but after smelling the fear sweat, the participants gave all three types the same attention (Social Cognitive and Affective Neuroscience, in press, DOI: 10.1093/scan/nsq097).

"It forced the brain to pay attention to things that otherwise it wouldn't consider worth its time," Mujica-Parodi says.

The smell of fear may be just one of many olfactory signals emitted by the human body.

Another study this year, by Yaara Yeshurun at the Weizmann Institute in Rehovot, Israel, and her team found that the imperceptible smell of women's tears decreases sexual arousal in men (Science, vol 331, p 226).

"It's a way of giving power to females, to make men less attracted to them," she says. The role of scent, or pheromones, in sexual attraction remains controversial, however.

The surprising thing about these studies is that few of the subjects were aware of the smells that they were facing, yet their behaviour was altered nevertheless.

The question, then, is why we pay so little conscious attention to our noses unless we get a whiff of something truly pungent?

Lee Sela and Noam Sobel, also at the Weizmann Institute, blame our obliviousness on two factors.

Firstly, they point out that our noses just aren't equipped to locate the source of an odour.

This makes the sense of smell fundamentally different to vision or hearing, which are built to pinpoint sights and sounds and turn them into a mental map.

According to one leading theory of consciousness, we become aware of something when the brain's "attentional spotlight" focuses on a single location, after which it picks out the fine details, like a familiar face, from the scene.

With such a poor map of smells, the spotlight can't shine on any particular part of the smellscape and make us aware of the details, say Sela and Sobel.

It's for this reason that we can only ever pick out around four smells from a complex mixture, they say.

The other reason centres on a phenomenon called change blindness, which was first found to influence our vision. In 1999, Kevin O'Regan from the Laboratory for the Psychology of Perception in Paris, France, and colleagues found that people can miss even large changes to a visual scene when those changes are accompanied by an interruption, such as a camera cutting to a different viewpoint in a film (Nature, vol 398, p 34).

They argued that the cut provides a fleeting distraction which means the change goes unnoticed.

 Since then, change blindness has been demonstrated in hearing and in touch.

Sela and Sobel think that smell could be next on the list.

They point out that our sniffs are full of gaps as we breathe in and out, which could make it difficult for us to notice new odours wafting around - even if we do react to them subconsciously (Experimental Brain Research, vol 205, p 13).

It's an interesting idea, says O'Regan, but he's not yet convinced. In particular, he is critical of the suggestion that sniffing more quickly would dissipate the effect.

"Even if you were going to sniff very quickly you would still have a break between each sniff." In visual change blindness, even the subtlest of cuts can mask large changes, he says.

There are other ways that we can improve our noses, though.

"We all have the capacity to train our sense of smell," says Dodd, "but you have to work at it." Master perfumers, for instance, learn to recognise, name and imagine an extraordinary range of smells through years of training.

This is accompanied by a significant reorganisation of the olfactory areas that helps them to process the scents more efficiently (Human Brain Mapping, in press, DOI: 10.1002/hbm.21207).

Jess Porter and colleagues at the University of California, Berkeley, have also been trying to train people's noses.

They persuaded 32 students to wear blindfolds and ear defenders, and get down on all fours to see whether they could sniff out a trail of chocolate oil.

Intrigued, I wanted to try it for myself, which is how I ended up on all fours in a London park.

My first attempt didn't go well - there seemed to be so many smells competing for my attention, including damp soil, grass and cigarette butts.

But, like the majority of the participants in Porter's experiment, I did manage to follow the trail to the end on the subsequent attempts, even when it deviated off a straight path.

For Porter's volunteers, repeated practice over three days brought greater accuracy and speed.

Of course you don't have to crawl on the grass to train your nose.

Any attempt to consciously pay attention to what your nose is telling you should have some benefit.

And even if you choose to ignore it entirely, there's no getting away from the fact that, behind the scenes, your nose is working overtime to make you who you are.

That's one discovery that's not to be sniffed at.

Half a century after our first and only visit, humans are returning to the deepest point in the ocean to uncover its secrets

IT IS a sobering fact that more people have walked on the moon's surface than have visited Earth's lowest spot. During six Apollo missions between 1969 and 1972, 12 humans did the moonwalk. Just two have plumbed Earth's ultimate depth. Their names are Jacques Piccard and Don Walsh, and their expedition to the Challenger Deep - the deepest point of the Mariana trench, some 11,000 metres below the surface of the Pacific Ocean near the island of Guam - lies over 50 years back, in January 1960.

The immense pressures and extreme cold of the ocean deep make reaching it a technologically demanding business. But sending humans into space isn't exactly easy. The discrepancy is largely down to politics: while the space race was fuelled by the rivalries of the cold war, no such spur has ever existed to encourage exploration of our oceans.

With the will to fund such ventures from the public purse on the wane, at least in the US and Europe, various private initiatives claim to be poised to take over the baton in the space race. Meanwhile something is stirring in the oceans too. The US film-maker James Cameron plans to send a crewed submersible into the Mariana trench to film footage for his follow-up to the film Avatar. And earlier this year, the British entrepreneur Richard Branson launched his one-person Virgin Oceanic submarine with the goal of visiting the deepest points of all of Earth's five oceans, the Challenger Deep included (see diagram). It is currently doing training runs with a view to performing the first full dives before the end of the year.

Branson is famous for his ability to attract publicity with well-funded, well-branded derring-do. "I didn't really understand why the oceans weren't being explored - it seems crazy when you think about it," he says. But this isn't just a Jules Verne-style yarn. A cadre of researchers have signed up to the project, and they have a long list of questions they want answered, from the ecology of deep-sea trenches to the role of trenches in Earth's geology. Is the light of science finally about to shine on our planet's deepest places?

The last time humanity made it down into the Mariana trench, it was ensconced in a contraption a little like a hot-air balloon in reverse. Piccard and Walsh's US navy submersible, Trieste, sank under its own weight into the trench and then discarded tonnes of iron shot to float back up. It spent 20 minutes at the bottom of the Challenger Deep, during which time the pair ate chocolate bars and looked out of the window. As they did so, they saw a shrimp-like creature float by in the inky blackness - a first proof that life could survive in a world with pressures well over 1000 times those at sea level.

In 1995, Japan's uncrewed Kaiko submersible provided further tantalising evidence of life down there in the shape of photographs of a worm, a sea cucumber and some shrimp. In May 2009, the uncrewed Nereus submersible, operated by researchers at the Woods Hole Oceanographic Institution in Massachusetts, collected liquid and rock samples from the bottom. And that's it - the sum total of humankind's interactions with the ocean's deepest depths.

The Mariana trench is a narrow scar in Earth's surface some 2500 kilometres long and averaging 70 kilometres wide. It began to form by the process of subduction some 50 million years ago, as the Pacific tectonic plate began to dive under the smaller Mariana plate to the west. The trench's depth is such that the currents that shuffle organisms around the rest of the ocean floor do not penetrate to its bottom. That makes the Mariana, and other trenches like it, evolutionarily isolated.

Charles Darwin showed 150 years ago how a similar isolation led the fauna of the Galapagos Islands to take bizarre forms - tortoises that grew to huge sizes, for example - and created the subtle diversity of Darwin's finches, each perfectly adapted to its island niche. But islands are mostly hospitable places, and can be reached by flying organisms, spores and pollen. The crushing pressures, cold temperatures and utter darkness of the ocean trenches, meanwhile, make them lethal to most creatures.

That means life found there was probably present as the trenches first started to form, allowing it to slowly adapt to the changing environment. "It really makes us wonder whether there are lost worlds of microbial organisms down there," says Katrina Edwards, a marine biologist at the University of Southern California in Los Angeles who is working with the Virgin Oceanic team.

But studying the ecology of the deep ocean floor is not as simple as dropping a probe on a cable, trapping something and hauling it back up. Organisms that live so deep tend not to survive the journey to the surface. "The pressure change just kills a lot of them," says Edwards. Instead, she and her colleagues rely on expensive automated landers, which use water pumps, filters and lures to collect principally microbial life and analyse it in situ.

Such kit is impressive, but limited. Visibility in trenches is poor and even with a live video feed the environment around the probe is generally unknown. "It is often hard to tell whether we have landed it on a cliff, in a ditch or near a vent," says Edwards. This matters: we know from terrestrial ecology that biological activity and diversity vary hugely according to local geology and chemistry. In subduction zones such as the Mariana trench, geological activity gives rise to hydrothermal vents and undersea volcanoes that create substantial differences in chemistry and temperature over small distances.

This is where Virgin Oceanic could prove invaluable. Edwards and her team will first drop a series of landers to the bottom of Challenger Deep, each equipped with analysis equipment including microscopes and DNA sequencers, as well as lures to attract and capture any creatures swimming nearby. Then, in a 2-hour swim along the bottom, the project's submersible will survey the areas around the landers, supplying sonar and video feeds of the topography around. "Our goal will be to visit each lander," says Chris Welsh, who will pilot the submersible. Mass spectrometers on board will also look for chemicals such as amino acids that are associated with life. "The hope is that continuously sampling the water flow will reveal active life areas that the vehicle has overflown," says Welsh.

It is not just biologists who have a keen interest in trench topography. Patricia Fryer, a geologist at the University of Hawaii in Honolulu, is hoping it will help us work out how Earth's continents first formed.

According to one theory, Earth's first dry land was created early on in our planet's history, when undersea subducting plates forced up overlying rock that eventually reached the sea surface as island arcs. Over geological time, further subduction along fault lines coupled with other tectonic movements caused these island arcs to migrate, collide and fuse, creating ever-larger land masses.

If this was the case, then patterns of elements and isotopes found in the middle of continents today, far away from fault lines, should also be present in active subduction zones. High levels of rubidium, strontium, barium, beryllium and light rare-earth elements found in continental interiors are thought to come from fluids driven off subducting tectonic plates and incorporated into the rocks of the overriding plate. Testing this idea means analysing sequences of rocks found in island arcs and down the deep fault scarps that lead out into the adjacent trenches.

Again, this is something we have so far done blind, dropping probes to the ocean floor with little knowledge of local factors that might skew the chemistry - a messy and expensive business. "A single experiment dropped in the wrong place can cost us millions of dollars," says Fryer. The plan for the Virgin Oceanic dive into the Mariana trench is to submerge near the island arc of Guam and travel slowly down the slope towards the Challenger Deep, taking high-definition video that should help identify the best locations for sampling stations. "It should prove a treasure trove for research," says Welsh.

One giant dive for humankind

The sudden release of forces built up along subduction zones as plates slip over and past one another also spawns great undersea earthquakes, often unleashing devastating tsunamis such as the one that hit the north-east coast of Japan earlier this year. Chemistry probably plays a significant part: as an ocean plate is scraped, pressurised and heated on its tortuous journey into Earth's interior, a patchwork of rocks of different strengths is created that could control when and where seismic activity is concentrated. "The more we know about these processes, the better we can understand where earthquakes are likely to occur," says Fryer.

The Mariana trench provides the perfect environment in which to do that, thanks to the mud volcanoes dotting its slopes. Mud volcanoes belch not fire, but fluids containing finely ground pieces of the subducting plate along with bits of the overlying rocks. Arranged at varying distances from the trench bottom, they provide an opportunity to tap material from as far as 20 kilometres down and so monitor the chemical processes occurring there.

On the Mariana trench dive, the Virgin Oceanic submersible is scheduled to visit a number of the mud volcanoes to record the coordinates of those that are active. That will not be without risk: the area is so geologically active that the explosive expulsion of material and hot water is a possibility, although Welsh anticipates that any serious activity will be picked up before the mission starts. The dangers posed by features such as overhangs and caves that could trap or damage the submarine are much more substantial, he thinks. Beyond some very sketchy sonar measurements taken from the surface, "these areas are not mapped at all", says Welsh.

Which leaves the obvious question, is it necessary for the Virgin Oceanic submersible to be crewed? Such a project to explore the ocean's deepest places could surely be carried out remotely, just like the two previous successful missions.

A crewed submersible might have some additional flexibility to respond in real time to unexpected points of interest on the ocean floor, but it is here, perhaps, where science meets showmanship. Far fewer people would have watched the moon landings had it not been an actual small step for a real representative of humankind. Branson is looking for a similar legacy. "I hope projects such as Virgin Oceanic will inspire generations who didn't witness the lunar landing to become great scientists of the future," he says. With an immense range of crucial answers to be fished out from the deep, a little pizzazz might go a long way.

Matt Kaplan is a freelance writer based in London and Los Angeles

Millions of ducks are flying south from Siberia this week, and some are carrying a virus that could lead to a resurgence of H5N1 bird flu in poultry – and people – across Eurasia.

"We're issuing an alert because we expect in the coming weeks to see the virus pop up in unexpected places across a wide area," says Jan Slingenbergh, head of the UN Food and Agriculture Organization's early warning system for animal diseases.

H5N1 has cost poultry farmers an estimated $20 billion so far.

It has also infected 565 people, of whom 331 died.

Virologists are trying to discover the mutations that could enable H5N1 to spread between people and go pandemic.

Dominant strain

The FAO is concerned about a strain of H5N1, called 2.3.2.1, which has been circulating for several years but is now emerging as dominant in birds in Asia.

It is no more virulent than previous strains but it is well adapted to many wild migratory species, so the virus has been carried to countries where H5N1 had been eliminated from poultry, including Bulgaria, Romania and Israel.

H5N1 outbreaks in poultry peaked in 2006, with 4000 across Eurasia and Africa. Extensive culling and vaccination quelled the virus, and by 2008 there were just 302 outbreaks.

With the spread of 2.3.2.1, outbreaks are back on the rise.

Some samples of the strain in China and Vietnam show its continuing evolution.

"The more cases you get in birds, the more it might spill over into humans," says Slingenbergh.

WORRIED about loose-tongued friends sharing your private details with the world?

Culling the least discreet members of your social network will help you feel more secure, but it's not a perfect solution.

What if your best friend is an offender?

Google's social networking site, Google+ had been running for less than a week when it turned out there was nothing to stop your friends "resharing" posts with the entire internet.

Google now lets users disable reshares, but the problem is indicative of how little control you have over what your friends do.

Pritam Gundecha at Arizona State University in Tempe has a technique for working out which friends are most likely to leak private information so you can remove them, if you choose.

Gundecha examined the relative importance of data 2 million Facebook users elect to share with the world and calculated the privacy risks friends pose to each other.

For example, around 80 per cent of users are happy to disclose their gender, but less than 1 per cent share their home address.

That suggests people publicising their address aren't particularly privacy-conscious and you might want to avoid them.

Using these statistics, the researchers gave each user a vulnerability score and worked out which friends will cause your vulnerability score to go down should you unfriend them.

It turns out that unfriending the least discreet friend increases your security by an average of more than 5 per cent - worth it for a casual acquaintance, but perhaps not so easy if your best buddy is a blabbermouth.

"There are some friends you cannot remove, irrespective of their vulnerability," admits Gundecha.

While the existing technique doesn't take this kind of social importance into account, he is now working on a version that does.

The preliminary work will be presented at the Knowledge Discovery and Data Mining conference in San Diego, California this week.

Randy Baden at the University of Maryland says unfriending people based on how vulnerable they make you is an intriguing take on the problem.

But he adds that the vulnerability scores are "based on how much potential there is for someone to leak information, not whether that person actually is leaking information".

A major roadblock in any negotiation can often be that each side believes their opponent's position is unmovable, according to a group of researchers in Israel.

Stand-offs, they say, can end if those involved think their counterparts can adopt a flexible mindset.

To put their theory to the test, they opted to ask Israelis and Palestinians.

Eran Halperin of the Interdisciplinary Center Herzliya in Israel and colleagues surveyed 500 Israeli Jews on whether they believed groups of people had fixed natures, and on their attitudes towards the Palestinians.

Volunteers who believed that groups could change tended to have more positive attitudes towards a negotiated peace process.

To find out if they could change those attitudes, the team then ran three experiments on 76 Israel-born Jews, 59 Israeli Arabs and 53 Palestinians living in the occupied Palestinian territories.

 In each case, volunteers were randomly asked to read one of two articles, which portrayed groups of people as having either a fixed or a flexible nature.

Jewish people were then asked about their attitudes towards Israeli Arabs and Palestinians and vice-versa, as well as how they felt about negotiating and compromising in a peace process.

Volunteers responded differently, depending on which article they had read: those who had read about flexible nature were more open to negotiations and compromises, when compared with those who had been given an article about inflexible human nature.

The difference between the two groups was statistically significant.

According to Halperin, this suggests that "when you make people believe that groups have malleable characteristics, they change their attitudes towards the other group and are more willing to make specific compromises for peace".

So are the findings useful in real world scenarios? Dominic Johnson, who studies conflict psychology at the University of Edinburgh, UK, is not convinced.

He says that while it is striking that people can be made to perceive other groups in a different way, Halperin's methods would be hard to implement.

The challenge in many conflicts is overcoming a small minority of committed, uncompromising extremists he says.

"They're the ones you ultimately have to defeat." Mind games may be of limited use there.

Journal reference: Science, DOI: 10.1126/science.1202925

THE tiny glass bottle in my hand is filled with what looks like crude oil, but it's actually oil's nemesis.

If it works, this black sludge will transform the rechargeable battery, doubling the range of electric cars and making petroleum obsolete.

Today's electric cars are handicapped by batteries that are heavy, expensive and a waste of space.

Two-thirds of the volume of the battery in Nissan's Leaf electric car, for example, consists of materials that provide structural support but generate no power.

And those materials cost more than the electrically active components.

One way to vastly improve rechargeable batteries is to put more of that deadweight to work.

That's the purpose of the secret sauce in the bottle, nicknamed "Cambridge crude" by Yet-Ming Chiang and his colleagues at the Massachusetts Institute of Technology, who developed it.

In a standard battery, ions shuttle from one solid electrode to the other through a liquid or powder electrolyte.

This in turn forces electrons to flow in an external wire linking the electrodes, creating a current.

In Chiang's battery, the electrodes take the form of tiny particles of a lithium compound mixed with liquid electrolyte to make a slurry.

The battery uses two streams of slurry, one positively charged and the other negatively charged.

Both are pumped across aluminium and copper current collectors with a permeable membrane in between. As they flow the streams exchange lithium ions across the membrane, causing a current to flow externally.

To recharge the battery, you apply a voltage to push the ions back across the membrane.

The MIT creation is a type of flow battery, which normally has a liquid electrolyte that moves past stationary electrodes.

Chiang reckons that the power per unit volume delivered by his lithium "semi-solid" flow battery will be 10 times that of conventional designs (Advanced Energy Materials, DOI: 10.1002/aenm.201100152).

"This is probably the most exciting development in electrical energy storage in the past couple of years," says Yury Gogotsi of Drexel Nanotechnology Institute in Philadelphia, Pennsylvania.

"Chiang offers a unique hybrid between a flow battery and a lithium-ion battery."

Drivers could have three ways of recharging the semi-solid flow battery.

They could pump out spent slurry and pump in fresh; head to a recharge station where tanks of spent slurry would be replaced with fresh ones; or recharge the slurries with an electric current.

In the first two cases regaining full power should only take a matter of minutes.

Rechargeable batteries are the heaviest and most expensive components of electric cars by a large margin.

Chiang estimates that the cost of manufacturing his team's battery will be $250 per kilowatt-hour of generating capacity.

So if one were built to replace the 24-kWh battery in the Nissan Leaf, it would cost $6000.

That is about one-third the cost of existing batteries, and just low enough to compete with gasoline.

Chiang also calculates that Cambridge crude would let a car travel at least 300 kilometres on a single charge, double what is possible with today's batteries.

"This is an especially beautiful technology," says Dan Steingart of the City University of New York Energy Institute, because you can recharge the spent slurry.

But he adds that even if the team manages to create a prototype car battery within five years, building the recharge stations to support it would take much longer.

Last year Chiang, his colleague Craig Carter and entrepreneur Throop Wilder founded a company called 24M Technologies to develop the battery.

They have raised $16 million in funding so far, and plan to have a compact prototype ready in 2013

A free drug can help treat many disorders with no side effects: our minds. New Scientist reveals six ways to exploit its power

"I TALK to my pills," says Dan Moerman, an anthropologist at the University of Michigan-Dearborn.

"I say, 'hey guys, I know you're going to do a terrific job'."

That might sound eccentric, but based on what we've learned about the placebo effect, there is good reason to think that talking to your pills really can make them do a terrific job.

The way we think and feel about medical treatments can dramatically influence how our bodies respond.

Simply believing that a treatment will work may trigger the desired effect even if the treatment is inert - a sugar pill, say, or a saline injection.

For a wide range of conditions, from depression to Parkinson's, osteoarthritis and multiple sclerosis, it is clear that the placebo response is far from imaginary.

Trials have shown measurable changes such as the release of natural painkillers, altered neuronal firing patterns, lowered blood pressure or heart rate and boosted immune response, all depending on the beliefs of the patient.

There is even evidence that some drugs work by amplifying a placebo effect - when people are not aware that they have been given the drugs, they stop working.

On the flip side, merely believing that a drug has harmful side effects can make you suffer them. The nocebo effect, as it's known, can even kill (New Scientist, 13 May 2009, p 30).

It has always been assumed that the placebo effect only works if people are conned into believing that they are getting an actual active drug.

But now it seems this may not be true. Belief in the placebo effect itself - rather than a particular drug - might be enough to encourage our bodies to heal.

In a recent study, Ted Kaptchuk of Harvard Medical School in Boston and colleagues gave some people with irritable bowel syndrome an inert pill.

They told them that the pills were "made of an inert substance, like sugar pills, that have been shown in clinical studies to produce significant improvement in IBS symptoms through mind-body self-healing processes", which is perfectly true.

Despite knowing the pills were inert, on average the volunteers rated their symptoms as moderately improved after taking them, whereas those given no pills said there was only a slight change (PLoS ONE, vol 5, e15591).

"Everybody thought it wouldn't happen," says study co-author Irving Kirsch, a psychologist at the University of Hull, UK.

He thinks that the key was giving patients something to believe in.

"We didn't just say 'here's a sugar pill'.

 We explained to the patients why it should work, in a way that was convincing to them."

As well as having implications for the medical profession, the study raises the possibility that we could all use the placebo effect to convince ourselves that sucking on a sweet or downing a glass of water, for example, will banish a headache, clear up a skin condition or boost the effectiveness of any drugs that we take.

"Our study suggests that might indeed help," says Kirsch.

While Moerman talks to his pills, Kirsch recommends visualising the desired improvement and telling yourself that something is going to get better.

Jo Marchant is a freelance writer based in London

SOME claim climate change will destroy our species; now it seems it also helped forge it.

The rapid fluctuations in temperature that characterised the global climate between 2 and 3 million years ago coincided with a golden age in human evolution.

The fossil record shows that eight distinct species emerged from one hominin species, Australopithecus africanus, alive 2.7 million years ago.

The first members of our genus appeared between 2.4 and 2.5 million years ago, while Homo erectus, the first hominin to leave Africa, had evolved by 1.8 million years ago.

To work out whether climate had a hand in the speciation spurt, Matt Grove of the University of Liverpool in the UK turned to a global temperature data set compiled by Lorraine Lisiecki at the University of California, Santa Barbara.

Lisiecki analysed oxygen isotopes in the shells of fossilised marine organisms called foraminifera.

During glacial periods, the forams' shells contain more of the heavier of two oxygen isotopes, as the lighter one is preferentially accumulated in snow and ice rather than the ocean.

Grove found that the mean temperature changed suddenly on three occasions during the last 5 million years.

Each change was equivalent to the difference between glacial and interglacial temperatures - but none of these episodes coincided with the hominin "golden age".

What marked out this period was a greater range of recorded temperatures, suggesting it was a time of rapid but short-lived fluctuations in climate.

Grove says such conditions would have favoured the evolution of adaptability that is a hallmark of the genus Homo (Journal of Archaeological Science, DOI: 10.1016/j.jas.2011.07.002).

Grove says the classic survival traits of H. erectus, forged during this period of change, include teeth suited for generalised diets and a large brain - both of which should have been advantageous at a time of swift climate change.

It's what women have been telling men for decades: stimulating the vagina is not the same as stimulating the clitoris.

Now brain scan data has added weight to their argument.

The precise locations that correspond to the vagina, cervix and female nipples on the brain's sensory cortex have been mapped for the first time, proving that vaginal stimulation activates different brain regions to stimulation of the clitoris.

The study also found a direct link between the nipples and the genitals, which may explain why some women can orgasm through nipple stimulation alone.

The discoveries could ultimately help women who have suffered nerve damage in childbirth or disease.

The sensory cortex is a strip of brain tissue positioned roughly under where the band between a pair of headphones sits.

Across it, neurons linked to different body parts exchange information about the sensory information feeding into them.

This is often depicted as the "sensory homunculus", a distorted image of a man stretched across the brain, with his genitals lying next to his feet (click here).

The size of the body's parts show how much of the brain is dedicated to processing the sensory information from each body part.

The diagram was first published in 1951 after experiments conducted during brain surgery performed while the patients were conscious: the surgeon electrically stimulated different regions of the patients' brains and the patients reported the parts of their bodies in which they felt sensation as a result. But all the subjects were men. Until recently, the position of female genitalia on the homunculus had only been guessed at.

This changed last year when a team led by Lars Michels at University Children's Hospital in Zurich, Switzerland, used functional magnetic resonance imaging to confirm that the position of the clitoris on the homunculus was in approximately the same position as the penis in men.

 Barry Komisaruk at Rutgers University in Newark, New Jersey, and his colleagues have now used the same method to map the position of the clitoris, vagina and cervix on the sensory cortex as women stimulated themselves.

There, there and there

"This is hard proof that there is a big difference between stimulating those different regions," says Stuart Brody of the University of the West of Scotland in Paisley, UK, one of the researchers in the study.

Some have argued that women who derive pleasure from vaginal stimulation do so because their clitoris is being indirectly stimulated, but the current findings contradict this.

"They support the reports of women that they experience orgasm from various forms of stimulation," says Beverly Whipple, also of Rutgers University, who was not involved in the current study.

It's the nipples, stupid

Komisaruk also checked what happened when women's nipples were stimulated, and was surprised to find that in addition to the chest area of the cortex lighting up, the genital area was also activated.

"When I tell my male neuroscientist colleagues about this, they say: 'Wow, that's an exception to the classical homunculus,'" he says.

 "But when I tell the women they say: 'Well, yeah?'" It may help explain why a lot of women claim that nipple stimulation is erotic, he adds.

The next step is to map what other areas of the brain light up in response to clitoral and vaginal stimulation.

Komisaruk would also like to see what happens when the area that supposedly contains the G-spot is stimulated, as women in the current study just stimulated the front wall of the vagina generally.

The findings could also help women who have suffered nerve damage in childbirth or because of diseases like diabetes.

Michels has preliminary evidence that stimulating the clitoral nerve can improve symptoms of urinary incontinence, but says a proper understanding of how the nerve maps to the brain is needed to translate this into effective treatment.

Meanwhile, Komisaruk says that nipple stimulation could enhance genital sensation in women with nerve damage.

 "It could be a supplement for experiencing orgasm," he says.

Journal reference: Journal of Sexual Medicine, DOI: 10.1111/j.1743-6109.2011.02388.x

A live implant could kill the pain associated with slipped discs, a study in rats suggests.

Between 1.5 and 4 million Americans are waiting for surgery to fix a herniated spinal disc, but the relief provided from a synthetic implant is the best it's ever going to be "the minute you put it into the patient", says Lawrence Bonassar of Cornell University in Ithaca, New York.

Living tissue can grow and adapt, so may provide a better long-term solution, he says.

Bonassar's team used cells taken from sheep spines to build replicas of rat discs, and implanted them into the spines of rats.

The implanted discs stood up to pulling and compression like the original discs.

Crucially, they also improved with age, growing new cells and binding to nearby vertebrae in the six months after surgery.

Although the study was in rats, "it shows us what is possible", says Abhay Pandit at the National University of Ireland in Galway.

He adds that future studies will need to address the load borne by upright human spines.

Journal reference: Proceedings of the National Academy of Sciences, DOI: 10.1073/pnas.1107094108

ANTARCTICA is rising like a cheese soufflé: slowly but surely.

Lost ice due to climate change and left-over momentum from the end of the last big ice age mean the buoyant continent is heaven-bound.

Donald Argus of NASA's Jet Propulsion Laboratory in Pasadena, California, and colleagues used 15 years of GPS data to show that parts of the Ellsworth mountains in west Antarctica are rising by around 5 millimetres a year (Geophysical Research Letters, DOI: 10.1029/2011gl048025). Elsewhere on the continent, the rise is slower.

A faster rise has been seen in Greenland, which is thought to be popping up by 4 centimetres a year.

Ongoing climate change could be partly to blame: Antarctica is losing about 200 gigatonnes of ice per year, and for Greenland the figure is 300 gigatonnes.

Earth's continents sit on viscous magma, so the effect of this loss is like taking a load off a dense foam mattress.

But there is another possible contributor.

"The Earth has a very long memory," says Argus.

As a result, "there is also a viscous response to ice loss from around 5000 to 10,000 years ago going on".

Despite this effect, the known ice loss at both poles suggests that embedded in the local rises is a signal of current climate change - researchers just have to tease it out.

WHY are we here? Where did we come from?

According to the Boshongo people of central Africa, before us there was only darkness, water and the great god Bumba.

One day Bumba, in pain from a stomach ache, vomited up the sun.

The sun evaporated some of the water, leaving land. Still in discomfort, Bumba vomited up the moon, the stars and then the leopard, the crocodile, the turtle, and finally, humans.

This creation myth, like many others, wrestles with the kinds of questions that we all still ask today.

Fortunately, as will become clear from this special issue of New Scientist, we now have a tool to provide the answers: science.

When it come to these mysteries of existence the first scientific evidence was discovered about 80 years ago, when Edwin Hubble began to make observations in the 1920s with the 100-inch telescope on Mount Wilson in Los Angeles County.

To his surprise, Hubble found that nearly all the galaxies were moving away from us.

Moreover, the more distant the galaxies, the faster they were moving away.

The expansion of the universe was one of the most important intellectual discoveries of all time.

This finding transformed the debate about whether the universe had a beginning.

If galaxies are moving apart now, they must therefore have been closer together in the past.

 If their speed had been constant, they would all have been on top of one another billions of years ago. Was this how the universe began?

At that time many scientists were unhappy with the universe having a beginning because it seemed to imply that physics had broken down.

One would have to invoke an outside agency, which for convenience one can call God, to determine how the universe began.

They therefore advanced theories in which the universe was expanding at the present time, but didn't have a beginning.

Perhaps the best known was proposed in 1948, and called the steady state theory.

According to this theory, the universe would have existed for ever and would have looked the same at all times.

This last property had the great virtue of being a prediction that could be tested, a critical ingredient of the scientific method. And it was found lacking.

Observational evidence to confirm the idea that the universe had a very dense beginning came in October 1965, with the discovery of a faint background of microwaves throughout space.

The only reasonable interpretation is that this background is radiation left over from an early hot and dense state.

As the universe expanded, the radiation would have cooled until it is just the remnant we see today.

Theory backed this idea too.